diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--bce4c43.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--bce4c43.json
new file mode 100644
index 0000000000000000000000000000000000000000..38a412a8c01c5efbd416f333dcabc50a2f52bf9d
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--bce4c43.json
@@ -0,0 +1,225 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-privileged--bce4c43",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 0,
+  "reward": 0.25,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 101,
+   "sim_steps": 101,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 0.25,
+   "stages_reached": 1,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 5.91,
+   "bad_trajectory": 0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 122,
+   "agent_resets": 117,
+   "agent_replays": 8,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.987,
+   "peak_processes": 12,
+   "peak_disk_gb": 0.24,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-privileged--bce4c43/replay.mp4",
+  "duration_s": 1.04,
+  "fps": 25.0,
+  "frames": 26,
+  "sha256": "901d60ad0f3ba3cf704e823baf0f62b2eb847b76e08b16ae619e735b00fd9f55"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–1.00 s 的全部 26 帧及机器人局部放大图显示，机器人在近侧发球区站立，双臂调整持球、持拍位置，随后右臂在身前低处挥拍；全段没有羽毛球飞越球网并落入远侧对角区的过程。"
+     },
+     {
+      "t": 1.0,
+      "log_step": 211,
+      "text_zh": "观察：log step 211 对已保存 trajectory.npz 的两次独立回放均返回 stages=[release]、steps=101、success=false，并报告“the racket hit the feathers first, not the base of the shuttle”；这是基准回放裁判的输出，说明已放球，但首次拍球接触不合法。视频 0.92–1.00 s 的末段仍在近侧挥拍，与此终止阶段一致。"
+     },
+     {
+      "t": 1.0,
+      "log_step": 221,
+      "text_zh": "观察：log step 216 恢复原挥拍参数并重新生成 101 行交付文件，log step 218 核对两次回放末态哈希相同、仅回放耗时不同，log step 221 从最终交付目录再次运行生成脚本仍返回先触羽毛的同一失败；inputs/run.json 的 reward 也给出 deterministic=1、stages_reached=1/4。推断：会话中其他调参测试不属于本视频；本次合法发球入区从未完成，不能将原生放球阶段的 0.25 当成语义节点完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 0.92–1.00 s 可见机器人右手所持球拍在身前挥动；log step 211 的交付回放裁判确认球拍拍面实际先碰到羽毛球羽毛。候选：击打；现有接触方式无击打，不能以推/扫或压/擦代替。左手先持球后放开，再由右手经球拍接触球；不重复标注右手握拍链路。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–1.00 s 双脚持续支撑球场地面，机器人未跌倒；这里记录可见的站立支撑，不据远景断言毫米级脚位与离地时长达标。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": null
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "serve_in_target",
+   "t": 1.0,
+   "log_step": 211,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 0.92–1.00 s 机器人正在完成低位挥拍，尚无过网飞行；log step 211 的已保存文件回放裁判在 101 个控制步后以“球拍先碰羽毛而非球托”终止，仅接受 release 阶段。log step 216、218、221 确认最终重新生成的文件仍重现同一失败。因此断点属于合法发球入区的击球操作。"
+  },
+  "unmet": {
+   "serve_in_target": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 1.0,
+    "log_step": 211,
+    "evidence_zh": "观察：视频 0.92–1.00 s 右手球拍在机器人身前低处与放球后的羽毛球位置汇合，远景不足以区分球托和羽毛；log step 211 对交付文件的回放裁判明确报告首先碰到羽毛。推断：拍面与羽毛球球托的相对对位未实现，击中错误部位；已有拍球接触，不标没建立接触，也无依据细分为击球过早或过晚。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "source_zh": "日志",
+    "evidence_zh": "推断：机器人未能把实际挥拍路径对准球托。log step 175 的生成脚本以固定腰部和双臂关节目标执行开环放球、撤手、挥拍；会话诊断 log step 179 记录接触附近球拍中心 y≈0.048 m、羽毛球 y≈0.175 m，log step 205 同一原始参数加随挥后的测试仍报告先碰羽毛、位置差约 0.147 m。log step 210、216 重新生成最终文件后，同一失败由 log step 211、218、221 的交付回放核实；这些结果支持挥拍对位的控制问题，但不足以确定是逆解、跟踪还是接触设计的单一子类。"
+   }
+  ],
+  "gap_zh": "没有状态日志；640×360 的远景视频仅以 25 fps 记录，放大后仍无法直接分辨球托首次接触及毫秒级接触时长，先触羽毛依据已核实最终交付的基准回放裁判。会话位置诊断用于支持控制归因，不作为逐帧状态路径或视频计时依据；不能进一步确定控制子类。评分器未报告 DAG 的五项原生指标，均记 null。"
+ },
+ "summary_zh": "机器人完成放球并实际挥拍触球，但未完成“合法发球入区”，语义节点最终为 0/1。最终交付回放在击球时因球拍先碰羽毛而非球托终止，现象为击球对位不准，日志支持控制问题。无状态日志且远景无法分辨细微接触，具体首触部位依赖交付回放裁判，控制子类与五项原生指标无法确定。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标签与视频、log 一致：放球阶段达成（release，0.25），球拍先碰羽毛导致失败，serve_in_target 未达成；断点、位姿不准现象和控制根因均有 log 回放裁判与 step 179/205 的几何数据支持，未发现需要修改之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:54:32Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--bce4c43/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--bce4c43/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "b1a2e33424c1c8c270512479e9652625488bbb3e41491c89f392c6fa22b1ded5",
+   "video_sha256": "901d60ad0f3ba3cf704e823baf0f62b2eb847b76e08b16ae619e735b00fd9f55",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--du3sucw.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--du3sucw.json
new file mode 100644
index 0000000000000000000000000000000000000000..fd1429030f570d73b926c94398408efd0ac76cac
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--du3sucw.json
@@ -0,0 +1,211 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-privileged--du3sucw",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 298,
+   "sim_steps": 298,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 4,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 8.2,
+   "bad_trajectory": 0,
+   "metric_hit_height_m": 0.7396,
+   "metric_landing_x_m": 5.933,
+   "metric_landing_y_m": 1.554,
+   "metric_shuttle_angle_deg": 38.16,
+   "metric_shuttle_speed_m_s": 11.6684,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 115,
+   "agent_resets": 584,
+   "agent_replays": 9,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 3.747,
+   "peak_processes": 31,
+   "peak_disk_gb": 0.843,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-privileged--du3sucw/replay.mp4",
+  "duration_s": 3.0,
+  "fps": 25.0,
+  "frames": 75,
+  "sha256": "078d7e00d016c6d43845b1169ad65b7f26cc6986893d549550f33c2fcbc04be7"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.98,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 1.68,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–1.52 s，机器人在近侧发球区原地调整手臂和球拍；视频 1.56–1.72 s 左手从放球处移开，右手球拍向前上方挥动；视频 1.76–2.00 s 白色羽毛球离开机器人向上飞向球网，两脚保持地面支撑。远景不能直接分辨球托初触及手指接触细节。"
+     },
+     {
+      "t": 2.0,
+      "log_step": 197,
+      "text_zh": "观察：视频 1.92–2.08 s 羽毛球由近侧升至网顶上方并飞向远侧淡黄色目标区方向；视频 2.24–2.96 s 机器人仍站立、右手仍持拍。log step 197 中最终选中轨迹的原生裁判输出依次记录 release、hit、over the net、in，success=True、failed=None，击球最高处 0.7396 m、出球速度 11.6684 m/s、仰角 38.16°、过网净空 0.575 m。"
+     },
+     {
+      "t": 2.98,
+      "log_step": 197,
+      "text_zh": "推断：log step 196 选择最终候选，log step 197 的生成输出为 298 个控制步、仿真 2.98 s，故控制步长为 0.01 s；298×0.01=2.98 s 与视频约 3.00 s 的长度相符，且裁判仿真 1.690 s 的击球与视频 1.68–1.72 s 挥拍出球相符，仿真 1.990 s 的过网与视频约 2.00 s 相符。据此将最终控制步对应约视频 2.98 s，作为节点完成锚点；裁判实际记录仿真 2.973 s 首次落地区内。视频末帧 2.96 s 尚未显示触地，完成时刻依赖这一已核对的交付回放时钟，而非 agent 工作时间。"
+     },
+     {
+      "t": 2.98,
+      "log_step": 202,
+      "text_zh": "观察：inputs/run.json 的评分器字段 success=1、replay_success=1、deterministic=1、stages_reached=4/4，落点为 (5.933,1.554) m，处于 DAG 指定的接发球区。log step 199 验证最终文件重新生成后字节一致且为 298 行；log step 200 启动该文件的两次新进程回放，log step 202 输出两次成功且末态哈希均为 8ad0c87716c429273574da561b752e29645f02dedfe0e43cb6ae80d442ae5be9，与 log step 197 一致，因此采用最终交付结果，未采用此前 302 步候选。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "候选：击打。观察：视频 1.68–1.80 s 机器人右手持球拍前上挥，随后羽毛球离拍向上飞行；结合 log step 197 的原生 hit 事件，判定实际发生球拍击球。球拍与右手的握持链路不重复计数；左手此前持球并先行放开，球托初触由裁判事件补充确认。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–2.96 s 机器人双脚支撑球场地面，发球期间身体保持站立，未见移步或跌倒。远景不足以独立量出毫米级脚部位移。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": 0.7396
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": 11.6684
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": 38.16
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": 5.933
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": 1.554
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人完成唯一节点“合法发球入区”：视频约 1.7 s 挥拍将球向上击出，视频约 2.0 s 越网，最终交付轨迹的裁判确认四阶段全部完成，落点 (5.933,1.554) m 在目标区内。最终 298 步文件经两次新进程成功回放，评分器确认成功且确定性一致，无失败断点。无状态日志，远景无法核验毫秒级接触和脚部位移，视频末帧 2.96 s 早于裁判记录的首次落地，故约视频 2.98 s 的完成锚点及精细合法性依赖已核对的最终交付回放输出与评分器字段；oracle 为本运行自身，没有独立参考。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "成功运行，唯一节点 serve_in_target 终态 true，完成锚点按已核对的控制步映射取视频约 2.98 s（裁判首次落地 sim 2.973 s），与 grader 字段一致；contact_actual（球拍击打、腿/躯干支撑）合理，无失败断点，未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:47:36Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--du3sucw/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--du3sucw/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "33bff7f102240c4478b1d1ea0a2218ce9bb76f1f296a829415a350cbab27288a",
+   "video_sha256": "078d7e00d016c6d43845b1169ad65b7f26cc6986893d549550f33c2fcbc04be7",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--kicwndu.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--kicwndu.json
new file mode 100644
index 0000000000000000000000000000000000000000..0e2073bcb9c3f9af1729b260c4ec3fca9d17d72f
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--kicwndu.json
@@ -0,0 +1,211 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-privileged--kicwndu",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 193,
+   "sim_steps": 193,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 4,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 5.96,
+   "bad_trajectory": 0,
+   "metric_hit_height_m": 0.5789,
+   "metric_landing_x_m": 5.174,
+   "metric_landing_y_m": 1.529,
+   "metric_shuttle_angle_deg": 36.37,
+   "metric_shuttle_speed_m_s": 9.7762,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 23,
+   "agent_resets": 2,
+   "agent_replays": 6,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 11.286,
+   "peak_processes": 28,
+   "peak_disk_gb": 0.982,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-privileged--kicwndu/replay.mp4",
+  "duration_s": 1.96,
+  "fps": 25.0,
+  "frames": 49,
+  "sha256": "871e2641bf05ab8210b66e63add0c983e715fc036912ae6c049ef018d39a4d70"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [
+     [
+      1.93,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.64,
+      "log_step": null,
+      "text_zh": "观察：已逐帧检查视频 0.00–1.92 s。视频 0.00–0.48 s 机器人左手将白色羽毛球带到身前，右手球拍由竖起转至较低位置；视频 0.60–0.84 s 白色羽毛球离开左手下落，视频 0.84–0.92 s 机器人右臂挥拍后羽毛球转为向前上方飞行，双脚仍支撑地面。"
+     },
+     {
+      "t": 1.12,
+      "log_step": null,
+      "text_zh": "观察：视频 0.96–1.12 s 的逐帧放大图中，白色羽毛球向球网方向升高；视频 1.16–1.92 s 机器人保持站立并持拍，球网遮住后续球路和首次落地接触点。成功参考视频也使用同一远景视角，仅用于核对可见线索，未用其结果替代本次证据。"
+     },
+     {
+      "t": 1.92,
+      "log_step": 162,
+      "text_zh": "观察：log step 162 对最终保存的 trajectory.npz 运行官方回放，裁判返回 success=true、failed=null，阶段依次为 release、hit、over the net、in，执行 193 个控制步、仿真 1.93 s；击球高度 0.5789 m、出球速度 9.7762 m/s、仰角 36.37°、首次落点 (5.174,1.529) m，落在 DAG 指定区间内。log step 164 的再次回放末态哈希相同；本次 reward 同样给出 replay_success=1、deterministic=1 和全部相同指标。"
+     },
+     {
+      "t": 1.93,
+      "log_step": 162,
+      "text_zh": "推断：log step 162 的最终交付回放执行 193 个控制步共仿真 1.93 s，对应每控制步 0.01 s；结合视频 25 fps、49 帧及末帧视频 1.92 s，将完成时刻近似定位为视频 1.93 s。已用 log step 152 的交付脚本在控制步 75 开始挥拍（后续修改仅调整放球步和挥拍终点），对照视频 0.76–0.92 s 的实际挥拍核对该映射。合法落地与毫秒级击球条件由最终交付回放裁判和评分器确认，未声称画面直接拍到首次触地。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "候选：击打；视频 0.84–0.92 s 机器人右手握持球拍挥动，羽毛球随之向前上方飞出，最终交付回放裁判确认 hit。作用端为球拍拍面，对象为羽毛球；现有方式没有击打，不用推/扫或压/擦代替，也不重复标注握拍链路。此前左手持球并放开，随后才由球拍击球。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 0.00–1.92 s 机器人双脚支撑球场地面，呈现球、放球、挥拍期间保持站立；不从远景推断毫米级脚位移。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": 0.5789
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": 9.7762
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": 36.37
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": 5.174
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": 1.529
+  }
+ ],
+ "failure": null,
+ "summary_zh": "本次交付完成“合法发球入区”节点，最终进度 1/1；视频可见机器人左手放球、右手挥拍后羽毛球向前上方飞出，最终交付回放及评分器确认四阶段全部通过，落点为 (5.174,1.529) m。节点在视频约 1.93 s 完成，时间由最终回放的控制步与视频挥拍核对后推断；无状态日志，后段球路和首次落地被球网遮挡，球托初触及毫秒级接触条件依赖裁判结果确认。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "成功运行，单节点 serve_in_target 达成、final=true，时间 1.93 s 由控制步映射推断并已标明；接触标注（手持工具·其他·击打、腿/躯干·支撑）合理，未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:53:40Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--kicwndu/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--kicwndu/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "e2d62bf5cf9df4f6edb266181bf2395e240ab5d8f8bee4baf13f317801194839",
+   "video_sha256": "871e2641bf05ab8210b66e63add0c983e715fc036912ae6c049ef018d39a4d70",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--mbmlydq.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--mbmlydq.json
new file mode 100644
index 0000000000000000000000000000000000000000..ae33cd1d627ec5a364c44cc9dafe1e545ea9da1e
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--mbmlydq.json
@@ -0,0 +1,221 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-privileged--mbmlydq",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 0,
+  "reward": 0.25,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 143,
+   "sim_steps": 143,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 0.25,
+   "stages_reached": 1,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 6.23,
+   "bad_trajectory": 0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 145,
+   "agent_resets": 6847,
+   "agent_replays": 0,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 2.68,
+   "peak_processes": 17,
+   "peak_disk_gb": 0.239,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 1,
+   "session_log_from_monitor": 1
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-privileged--mbmlydq/replay.mp4",
+  "duration_s": 1.44,
+  "fps": 25.0,
+  "frames": 36,
+  "sha256": "c3e0adebe8b4c07d0dd6b338c601156bae469720e085e3bf1ce2082211b7275a"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": 13,
+      "text_zh": "观察：视频 0.00–1.00 s 机器人双脚站在近侧场地、右手持拍，未见挥拍；视频 1.04–1.36 s 白色羽毛球从左手附近向身前下方下落，球拍仍留在右手前上方，未见羽毛球向球网飞出。"
+     },
+     {
+      "t": 1.4,
+      "log_step": 13,
+      "text_zh": "观察：视频 1.40 s 末帧机器人仍站立持拍，球没有飞到远侧目标区；log step 13 保存的占位轨迹返回“you missed the shuttle: it reached the floor without a hit”，共 143 个动作，与本次 reward 的 n_actions=143、stages_reached=1/4、replay_success=0 一致。推断：本次只完成放球，未完成合法击球、过网和入区，事件节点全程未达成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.00–1.00 s 左手在身前持羽毛球，随后放开，视频 1.04–1.36 s 球向下离开手部；右手始终握球拍拍柄。视频未见球拍拍面击中羽毛球，故不填写手持工具击打接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–1.40 s 机器人双脚在近侧球场地面支撑身体，放球过程中保持站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": null
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "serve_in_target",
+   "t": 1.12,
+   "log_step": 13,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 1.04–1.36 s 左手放出的羽毛球在机器人身前下落，右手球拍没有挥向球，未发生向网击出的过程。log step 13 的保存结果明确为球未击中便落地，保存动作数 143 与评分记录一致；该次发球失败后没有修复。"
+  },
+  "phenomena": [
+   {
+    "cat": "接触",
+    "item": "没建立接触",
+    "t": 1.12,
+    "log_step": 13,
+    "evidence_zh": "观察：视频 1.04–1.36 s 羽毛球从手部附近下落，右手球拍保持在身体右前上方，没有随球下挥或击出球；log step 13 保存轨迹的反馈为“you missed the shuttle: it reached the floor without a hit”。所需的球拍与羽毛球击打接触未建立。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "log_step": 13,
+    "source_zh": "日志",
+    "evidence_zh": "推断：交付的动作序列未实现放球后的挥拍控制。log step 13 的保存脚本逐步复制初始 hold，仅在循环 k≥100 时把左手开合量 row[17] 改为 0，右臂目标保持不变；视频 1.04–1.40 s 的持拍不挥与此一致。后续会话 log step 50 的带裁判轨迹网格仍多次返回漏球或击球前停挥，log step 53 的动态逆解修正返回“no valid contact candidate”，支持挥拍轨迹控制未解决；这些测试不计入本次节点进度。"
+   }
+  ],
+  "gap_zh": "观察：inputs/ 没有状态日志；视频仅有远景，羽毛球很小，视频 1.40 s 末帧不能精确核实首次触地时刻及接触细节，落地判负由 log step 13 的已保存轨迹反馈与评分记录支持。五项原生指标在 reward 中均未报告，保留 null；没有足够证据进一步区分控制子类。log step 38 只写出含占位内容的生成模板，后续会话搜索没有已执行的重新保存记录，不能将其测试结果视为本次交付。",
+  "unmet": {
+   "serve_in_target": false
+  }
+ },
+ "summary_zh": "机器人在视频约 1.04 s 放开羽毛球后保持持拍站姿，球在身前下落，未建立拍球接触；合法发球入区为 0/1，原生裁判仅完成四阶段中的放球阶段。交付对应 log step 13 的 143 动作占位轨迹，其脚本只打开左手而未加入挥拍；后续会话轨迹搜索不属于本次交付。无状态日志且球体在远景中很小，精确触地时刻与原生速度、击球高度等指标无法核实。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标注与视频、log 一致：视频 0–1.44 s 机器人持拍不动，约 1.04 s 后左手放出小球下落，球拍没有挥动；log step 13 保存的是只开左手的占位轨迹（143 动作，与 n_actions 一致），之后会话从未重新保存。断点、现象（没建立接触）、事件节点未达成与 contact_actual 均有证据支持。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:57:15Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--mbmlydq/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--mbmlydq/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "1986223e4c4c8870549710fff95682f026ce9f748cbf7eedcdd754f88fc6134e",
+   "video_sha256": "c3e0adebe8b4c07d0dd6b338c601156bae469720e085e3bf1ce2082211b7275a",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--xhnfjzn.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--xhnfjzn.json
new file mode 100644
index 0000000000000000000000000000000000000000..3557ba5f64a32ca5ad75f60501b26062a3ce5f12
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--xhnfjzn.json
@@ -0,0 +1,209 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-privileged--xhnfjzn",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 196,
+   "sim_steps": 196,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 4,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 5.36,
+   "bad_trajectory": 0,
+   "metric_hit_height_m": 0.9299,
+   "metric_landing_x_m": 5.45,
+   "metric_landing_y_m": 1.536,
+   "metric_shuttle_angle_deg": 43.21,
+   "metric_shuttle_speed_m_s": 10.455,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 714,
+   "agent_resets": 277,
+   "agent_replays": 19,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-privileged--xhnfjzn/replay.mp4",
+  "duration_s": 2.0,
+  "fps": 25.0,
+  "frames": 50,
+  "sha256": "3c0b8e25d30ea1f05c91b3b69e9b7babdbda64492e103a2f59df198542f84bd3"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [
+     [
+      1.96,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00 s，机器人站在近侧发球区，右手持球拍，双脚着地；视频 0.00–0.52 s 调整持拍手臂，没有完成发球。"
+     },
+     {
+      "t": 0.64,
+      "log_step": null,
+      "text_zh": "观察：视频 0.56–0.72 s，机器人右臂带球拍向前上方挥动，白色羽毛球随后离开机器人向上飞行；视频 0.84–1.60 s，白球继续向球网另一侧飞行并下降。视频 1.64–1.96 s 球很小且逐渐受球网遮挡，机器人仍站立持拍；画面不能直接分辨首次触地。"
+     },
+     {
+      "t": 1.96,
+      "log_step": 432,
+      "text_zh": "观察：log step 430 明确对交付的 trajectory.npz 启动两次独立回放；log step 432 的基准回放裁判（仿真中的规则检查器）输出均为 success=true、failed=null、196 控制步、time_s=1.96，并具有相同末态哈希。其事件依次为 release（仿真 0.580 s，脱离接触起点仿真 0.550 s）、hit（仿真 0.612 s）、over the net（仿真 0.967 s，越网余量 0.918 m）、in 和 success（仿真 1.959 s）；落点为 (5.450, 1.536) m。inputs/run.json 的 reward 同样报告 success=1、replay_success=1、deterministic=1、stages_reached=4/4 及完全相同的五项原生指标。"
+     },
+     {
+      "t": 1.96,
+      "log_step": 432,
+      "text_zh": "推断：交付回放的 196 控制步对应仿真 1.96 s，即每步 0.01 s；视频以 25 fps 呈现该次短发球，视频 0.60–0.64 s 的挥拍、随后白球飞离与裁判仿真 0.612 s 的击球吻合。结合末帧视频 1.96 s，将仿真 1.959 s 的入区成功近似锚定在视频 1.96 s，事件此后锁存。首次落地及球托初触、毫秒级接触/脚步规则依据上述交付回放裁判与本次评分确认，并非从远景直接量得。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "候选：击打；现有方式没有瞬时击打，不能用推/扫或压/擦代替。观察：视频 0.56–0.72 s 机器人右手持拍前挥后白色羽毛球上升飞离。推断：结合 log step 432 的 release→hit 裁判事件，左手先放球，随后右手经球拍拍面击打羽球；持拍的握持链路不另计。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–0.64 s，机器人双脚在球场上承重并完成挥拍，随后至视频 1.96 s 仍以双脚站立。远景不能量出毫米级脚部位移或毫秒级离地间隙。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": 0.9299
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": 10.455
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": 43.21
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": 5.45
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": 1.536
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人完成一次放球、挥拍、越网并落入对角接发球区的合法发球，唯一事件节点在视频约 1.96 s 达成并锁存，最终为 1/1。交付文件的两次独立裁判回放与本次评分均成功，击球高度 0.9299 m、落点 (5.450, 1.536) m，失败标注为空。已逐帧检查完整视频及末段，并与成功参考画面对照；本次无状态日志，首次触地受球网遮挡，精细接触规则和入区结果以交付裁判及评分字段确认，达成时间为经挥拍画面核对后的近似视频锚点。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "单节点成功标注与视频、评分字段（4/4、196步=1.96 s）一致，接触与时钟无误，failure 为空合理。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:48:33Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--xhnfjzn/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--xhnfjzn/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "30e35e84da177f88f02dff8e9a3f3fb911e0132a5133b74c892016473799e3dd",
+   "video_sha256": "3c0b8e25d30ea1f05c91b3b69e9b7babdbda64492e103a2f59df198542f84bd3",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--xnwja8r.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--xnwja8r.json
new file mode 100644
index 0000000000000000000000000000000000000000..c09ec52c8aa13ce42fc1b4a6817e92a32d5b7930
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--xnwja8r.json
@@ -0,0 +1,231 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-privileged--xnwja8r",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 0,
+  "reward": 0.5,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 105,
+   "sim_steps": 105,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 0.5,
+   "stages_reached": 2,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 5.87,
+   "bad_trajectory": 0,
+   "metric_hit_height_m": 0.7667,
+   "metric_landing_x_m": 1.532,
+   "metric_landing_y_m": -0.057,
+   "metric_shuttle_angle_deg": 36.45,
+   "metric_shuttle_speed_m_s": 2.7081,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 129,
+   "agent_resets": 3063,
+   "agent_replays": 9,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.157,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.26,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-privileged--xnwja8r/replay.mp4",
+  "duration_s": 1.08,
+  "fps": 25.0,
+  "frames": 27,
+  "sha256": "7c83250e8029c82e87e3dbbad245d5e3c75811f3fdc444ae24aba89ac80962c0"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.44,
+      "log_step": 393,
+      "text_zh": "观察：视频 0.28–0.44 s，机器人左手附近的白色羽毛球下落，右手球拍从低处向前上方挥动；视频 0.48–0.64 s 羽毛球离开机器人向前短飞。log step 388 对最终交付文件重新生成并回放，log step 393 的基准裁判输出确认 release、hit 两阶段，击球高度 0.7667 m、出球速度 2.7081 m/s、仰角 36.45°，与评分器 reward 字段相同。"
+     },
+     {
+      "t": 1.04,
+      "log_step": 393,
+      "text_zh": "观察：逐帧检查视频 0.00–1.04 s；视频 0.84–1.04 s 白球始终在网前近侧并下降至接近地面，没有越网进入远侧淡黄色目标区。log step 393 的最终交付回放裁判判为 the service landed on your own side，落点 x=1.532 m、y=-0.057 m，只有 release、hit，没有 over the net、in；reward 的 success=0、deterministic=1，落点与该回放一致。"
+     },
+     {
+      "t": 1.04,
+      "log_step": null,
+      "text_zh": "推断：结合视频末段网前下降的飞行和交付回放裁判落点，球未到 x∈[4.43,9.15] m、y∈[0.43,3.04] m 的接发球区，完整合法发球节点全程未达成。成功参考视频 1.00–2.96 s 中白球可飞越网并进入远侧区域，只用于比较视野中的完成线索。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 0.40–0.48 s 机器人右手持球拍向前上方挥动，随后白色羽毛球向前飞出；log step 393 的交付回放裁判确认球拍击球。候选：击打；现有方式没有瞬时击打，故用其他。左手此前持球并张开放球，握拍链路不重复计。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–1.04 s 机器人双脚在场地原位支撑身体，放球和挥拍期间没有可见移步或跌倒。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": 0.7667
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": 2.7081
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": 36.45
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": 1.532
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": -0.057
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "serve_in_target",
+   "t": 0.44,
+   "log_step": 393,
+   "where": "操作中",
+   "evidence_zh": "推断：视频约 0.44 s 的挥拍把羽毛球送成短而低的飞行，视频 0.84–1.04 s 球仍在己方网前下降，之后没有修复机会。log step 388–393 对最终交付轨迹的回放明确判为己方落地，故断点是正在执行的合法发球入区步骤。"
+  },
+  "unmet": {
+   "serve_in_target": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 0.44,
+    "log_step": 393,
+    "evidence_zh": "观察：视频 0.48–1.04 s，机器人击出的白球只在己方网前短距离飞行并下落，没有进入对面接发球区；log step 393 的交付回放裁判记录落点为 (1.532,-0.057) m，判为己方落地。推断：发球飞行距离不足、落点偏短，属于击出后的目标位置未到达。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "log_step": 317,
+    "evidence_zh": "推断：agent 的交付生成脚本采用固定的后摆、左手放球和余弦前挥关节指令（log step 317），其击球动作产生的出球速度及方向不足以把球送过网到目标区。最终交付回放（log step 393）已确认合法 release、hit，但出球仅 2.7081 m/s、仰角 36.45°，落点仍在己方；视频 0.48–1.04 s 的低而短飞行支持挥拍控制未实现所需发球轨迹。现有证据不能进一步区分拍面接触几何、关节轨迹或具体参数问题。",
+    "source_zh": "日志 / 视频 / 评分器字段"
+   }
+  ],
+  "gap_zh": "没有状态日志。视频为 25 fps 远景，能看见挥拍、球飞出及末段网前下降，但无法独立核实球托初触、毫秒级接触时长、脚部微移和首次触地瞬间；精细合法性与落点采用与 reward 一致的最终交付回放裁判记录（log step 393）。未将日志中的仿真事件时刻直接写作视频时刻，断点仅取已检查的视频约 0.44 s；控制根因的具体子类证据不足。"
+ },
+ "summary_zh": "机器人完成放球和击球两个原生阶段，但视频中羽毛球仅在己方网前短飞后下落，合法发球入区节点未完成（0/1）。最终交付回放裁判与评分器均确认己方落地，断点为发球操作中的出球轨迹偏短，根因归于控制。无状态日志且视频为远景，精细接触与首次落点依赖交付回放裁判，不能细分控制子类。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点未达成、断点 serve_in_target、位姿不准现象与控制根因均有视频帧及 log step 393 的交付回放裁判（stages release/hit，落点 1.532,-0.057，己方落地）支持；时钟均为视频时间，未发现需改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:50:16Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--xnwja8r/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--xnwja8r/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "c5d50a1feaf655a116f1ba713df24b38e9c3a90d8cf934646e8ded4bfa9107b4",
+   "video_sha256": "7c83250e8029c82e87e3dbbad245d5e3c75811f3fdc444ae24aba89ac80962c0",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--xx6kkxy.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--xx6kkxy.json
new file mode 100644
index 0000000000000000000000000000000000000000..1fa5d1937476abe102076f20580f40c18d1a6f99
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-privileged--xx6kkxy.json
@@ -0,0 +1,211 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-privileged--xx6kkxy",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 569,
+   "sim_steps": 569,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 4,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 9.9,
+   "bad_trajectory": 0,
+   "metric_hit_height_m": 0.8135,
+   "metric_landing_x_m": 5.227,
+   "metric_landing_y_m": 1.457,
+   "metric_shuttle_angle_deg": 33.69,
+   "metric_shuttle_speed_m_s": 9.6682,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 211,
+   "agent_resets": 2606,
+   "agent_replays": 2,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 4.177,
+   "peak_processes": 26,
+   "peak_disk_gb": 0.244,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-privileged--xx6kkxy/replay.mp4",
+  "duration_s": 5.72,
+  "fps": 25.0,
+  "frames": 143,
+  "sha256": "c05026e52e74142cd03ce7dae8b5161600438c6d200712a37fef75ab26263d01"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.69,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.52,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–4.28 s，机器人站在近侧发球区，调整左臂持球位置及右手球拍姿态；视频 4.32–4.64 s 左手回撤，右臂挥拍，视频 4.68–4.88 s 白色羽毛球与球拍分离后向上、向球网方向飞行。已逐帧检查视频 4.28–4.92 s；接触瞬间的球托朝向与毫秒级时长不能由该远景直接分辨。"
+     },
+     {
+      "t": 4.92,
+      "log_step": 96,
+      "text_zh": "观察：视频 4.84–4.92 s，白色羽毛球到达网顶上方并飞向远侧；视频 4.96–5.68 s 可见其继续朝远侧对角区域下落，机器人仍站立且球拍随右手运动。log step 96 中，agent 生成并保存最终提交后执行两次官方回放；仿真裁判（基准内判定发球阶段的程序）的输出均为 success=True、failed=None、release→hit→over the net→in，且末态哈希相同，与评分器 reward 的 replay_success=1、deterministic=1 和四阶段通过一致。"
+     },
+     {
+      "t": 5.69,
+      "log_step": 96,
+      "text_zh": "推断：节点在视频约 5.69 s 达成并锁存。log step 95 的生成器按每行 0.01 s 控制，log step 96 对所交付文件的两次正式回放均为 569 控制步，即仿真 5.69 s，与视频 5.72 s 的长度及末帧视频 5.68 s 一致；交付输出的过网事件为仿真 4.894 s，已在视频 4.88–4.92 s 的球越网处核对该时钟对应关系。该输出在仿真 5.689 s 记录 in 和 success，评分器报告首次落点 (5.227,1.457) m，落在 DAG 目标区内，因此以视频约 5.69 s 为达成锚点；末帧视频 5.68 s 未清楚显示实际触地，不把落地瞬间称为直接视觉观察。"
+     },
+     {
+      "t": 5.69,
+      "log_step": 96,
+      "text_zh": "观察：inputs/run.json 的评分器字段给出 success=1、stages_reached=4、stages_total=4、fell=0，击球高度 0.8135 m、出球速度 9.6682 m/s、仰角 33.69°、落点 x=5.227 m、y=1.457 m；log step 96 的最终提交正式回放报告相同指标。放球脱离时长、球托初触、脚步微小位移及接触持续时间由裁判通过结果支持，远景视频不能逐项独立量测。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "候选：击打；视频 4.52–4.72 s 机器人右手所持球拍上挥，羽毛球随后上升飞离，log step 96 最终提交回放的裁判确认 hit。球拍拍面瞬时作用于羽毛球，没有适用的既有接触方式，故用其他；此前左手持球并在视频约 4.32–4.44 s 回撤放球，右手握拍链路不另重复计为抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 0.00–4.64 s 可见机器人双脚在近侧球场支撑身体完成准备及挥拍，视频 4.68–5.68 s 身体略有转动但仍保持站立；双脚的毫米级移动与极短离地无法由远景量测。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": 0.8135
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": 9.6682
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": 33.69
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": 5.227
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": 1.457
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人完成左手放球、右手持拍上挥和羽毛球越网飞向远侧对角区域；最终提交的裁判回放于视频约 5.69 s 确认首次落点入区，唯一节点“合法发球入区”达成，最终进度 1/1。评分器判成功且确定性复现，击球高度 0.8135 m、落点 (5.227,1.457) m 均符合相应阈值。无状态日志，远景无法独立核验毫秒级接触及脚步细节，末帧视频 5.68 s 也未清楚拍到触地瞬间，这些验收由交付回放的裁判事件和评分器指标支持。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "成功运行，唯一节点 serve_in_target 达成并以交付回放的 569 控制步（仿真 5.69 s）映射到视频约 5.69 s，已在越网事件处核对时钟；contact_actual（手持工具·其他·击打、腿/躯干·支撑）与评分器指标一致，无失败层，未发现需修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:53:25Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--xx6kkxy/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-privileged--xx6kkxy/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "99670daa365d815cab578285f2ea0a5b83c0d31bb84a894444a85e8f429b87e0",
+   "video_sha256": "c05026e52e74142cd03ce7dae8b5161600438c6d200712a37fef75ab26263d01",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--5atxlb5.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--5atxlb5.json
new file mode 100644
index 0000000000000000000000000000000000000000..a5b021e6e8e145e55cd210622d822fea1e7d14cc
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--5atxlb5.json
@@ -0,0 +1,229 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-standard--5atxlb5",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 193,
+   "sim_steps": 1821,
+   "calls": 67,
+   "rows": 912,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2525.0,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-standard--5atxlb5/replay.mp4",
+  "duration_s": 1.96,
+  "fps": 25.0,
+  "frames": 49,
+  "sha256": "f3a7ee814671016637ce8dc4e7ae43cb4b38c9955bc03a1ab4b4fab4ad319dbf"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.6,
+      "log_step": 192,
+      "text_zh": "观察：视频 0.00–1.60 s，机器人双脚站在近侧场内，左臂将羽毛球举到身前，右手持拍调整并开始挥动；log step 192 的最后一次 reset 后，服务在准备结束时报告 holding_left=1.0、holding_right=1.0。"
+     },
+     {
+      "t": 1.92,
+      "log_step": 192,
+      "text_zh": "观察：逐帧检查视频 1.64–1.92 s，机器人左手撤回、右手球拍向上挥过，白色羽毛球仍在机器人身前下落，没有向球网及远侧目标区飞行；末帧机器人仍站立、右手仍持拍。log step 192 的第六局服务返回 done=True、stages_reached=1，裁判原文为“the racket's forward movement stopped before the hit”；log step 194、198 均确认最后一局已经失败且未再重置。"
+     },
+     {
+      "t": 1.92,
+      "log_step": 192,
+      "text_zh": "推断：本次评分局只完成放球，未完成击球、过网和合法首次落地，因此“合法发球入区”全程未达成；此判断结合可见挥空、球留在近侧及本局裁判结果，并非仅由 success=0 得出。成功参考视频 1.80–2.96 s 可见球离开机器人飞向远侧，该视觉结果在本次视频中没有出现。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.00–1.60 s，左手抓持羽毛球并抬臂定位，右手抓握球拍拍柄并调整拍姿，两臂各持一物配合准备发球；log step 192 的准备结束读数也显示两手 holding 均为 1.0。随后左手张开放球、右手持拍挥动，裁判未确认拍球击打，故不把计划中的手持工具击打记作实际接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–1.92 s，机器人双脚在球场地面支撑站立，发球动作中没有可见移步或跌倒；没有状态日志，不能核验毫米级脚部位移或短暂离地。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "value": null,
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better"
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "value": null,
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better"
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "value": null,
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better"
+  },
+  {
+   "name": "landing_x_m",
+   "value": null,
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target"
+  },
+  {
+   "name": "landing_y_m",
+   "value": null,
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "serve_in_target",
+   "t": 1.9,
+   "log_step": 192,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 1.64–1.92 s，机器人在放球后挥拍，羽毛球未被向网击出而在身前下落；log step 192 的最后一局裁判在仿真 1.93 s 报告“the racket's forward movement stopped before the hit”、stages_reached=1、done=True。失败发生在发球动作内，后续 log step 194 的重置被拒绝，本局没有修复。"
+  },
+  "unmet": {
+   "serve_in_target": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 1.9,
+    "log_step": 192,
+    "evidence_zh": "推断：视频 1.64–1.92 s 的右手上挥与左手放出的下落羽毛球没有形成击球时刻的相遇，球仍在身前而球拍已挥至较高处；log step 192 的本局裁判明确报告击球前前挥停止，阶段停在放球。这里是对运动中羽毛球的挥空，不另将同一事件重复标为没建立接触。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "log_step": 192,
+    "evidence_zh": "推断：log step 192 的实际执行脚本用预设右臂关节插值挥拍，并按固定行序打开左手、撤回左臂，挥拍段未依据羽毛球观测重新调整；同一步服务报告击球前前挥停止，视频 1.64–1.92 s 可见球与球拍错过。证据支持放球与挥拍的控制配合未实现，但不能区分接触几何、实际关节跟踪或具体时间参数哪项主导，故保留在“控制”。",
+    "source_zh": "日志"
+   }
+  ],
+  "gap_zh": "无状态日志；远景中羽毛球及手指很小，不能量出球拍与羽毛球的三维最近距离、毫秒级接触和实际速度，也不能分离挥拍轨迹误差与放球时刻误差。log step 192 的裁判结果属于最后评分局，可支持未击球的判断；前五局的脱拍、触身或落地记录均不作为本局现象。DAG 所列五项原生指标未在本次评分器 reward 中报告，均保留 null。"
+ },
+ "summary_zh": "本次评分是最后一次重置后的第六局，机器人站立持拍、举球并放球挥拍，但球未被击向球网，“合法发球入区”最终为 0/1。视频末段可见羽毛球仍在身前下落，log step 192 的本局裁判报告击球前球拍前挥停止，断点为发球操作中的动态时机失配，根因保留为控制。缺少状态日志和精细拍球运动数据，无法再区分轨迹、跟踪与放球时间参数的责任。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "节点未达成、断点 serve_in_target、末局（第六局）视频 1.96 s 与 log step 192（ready 1.2 s、挥拍结束 sim 1.93 s、stages=1、裁判“前挥在击球前停止”）一致；现象（动态时机）与根因（控制，未细分）有证据支持且标为推断，未发现需要改的内容。球在远景中极小，视频只能佐证未飞向球网。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:56:57Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--5atxlb5/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--5atxlb5/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "149251bc83ea81eb77dc3f90d65721b4ad151bcda3d0bbfc9e94b7c904f4b476",
+   "video_sha256": "f3a7ee814671016637ce8dc4e7ae43cb4b38c9955bc03a1ab4b4fab4ad319dbf",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--cshxrw7.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--cshxrw7.json
new file mode 100644
index 0000000000000000000000000000000000000000..8bee7f25559fb88379ea85acf308a451b2691796
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--cshxrw7.json
@@ -0,0 +1,230 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-standard--cshxrw7",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 473,
+   "sim_steps": 2592,
+   "calls": 291,
+   "rows": 2588,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 1930.0,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.172,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.329,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-standard--cshxrw7/replay.mp4",
+  "duration_s": 4.76,
+  "fps": 25.0,
+  "frames": 119,
+  "sha256": "ed985af1cc064ce689ef4b809aade8ef86fa04a28249a1094ca235294193c172"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 88,
+      "text_zh": "观察：视频 0.00–1.60 s，机器人在近侧球场站立，左手持白色羽毛球调整位置，右手将球拍降到身体右侧；视频 2.00–4.32 s 继续调整左臂，双脚支撑地面。log step 88 开始最后一次重置后的尝试；log step 102 的服务状态确认最终为 episode 6，control_steps=473。"
+     },
+     {
+      "t": 4.48,
+      "log_step": 101,
+      "text_zh": "观察：视频 4.44–4.60 s，白色羽毛球由左手旁向下移动；视频 4.68–4.72 s，机器人右手球拍开始向前挥动，末帧羽毛球仍在机器人近旁，未见越网或在远侧淡黄色目标区落地。log step 101 的服务裁判返回 done=true、reason=failed、stages_reached=1/4，明确指出“the racket hit the feathers first, not the base of the shuttle”；这是服务裁判的击球接触判定，不是 agent 自述。"
+     },
+     {
+      "t": 4.72,
+      "log_step": 101,
+      "text_zh": "推断：本次只完成原生放球阶段，击球因先触羽毛被裁判终止，因此整个“合法发球入区”事件从未成立。log step 3 给出控制周期 0.01 s，log step 101 的末控制步 473 对应仿真 4.73 s，与视频 4.68–4.72 s 的挥拍及视频总长 4.76 s 一致；断点以已检查的视频 4.72 s 为近似锚点，未将 agent 工作时间换算为视频时间。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 4.68–4.72 s 机器人右手握持球拍前挥；log step 101 的服务裁判确认球拍实际先击中羽毛球的羽毛。候选：击打；现有方式没有击打，不能用推/扫或压/擦替代。此前视频 0.00–4.40 s 左手持球、随后张开放球；右手握拍链路不重复记接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–4.72 s 机器人双脚在近侧球场地面承重，调整双臂并挥拍时仍站立，未见身体倒地。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": null
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "serve_in_target",
+   "t": 4.72,
+   "log_step": 101,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 4.44–4.60 s 羽毛球从左手旁下落，视频 4.68–4.72 s 机器人挥动右手球拍；log step 101 中，服务裁判在本次第 473 控制步判失败，原因是球拍先触羽毛而非球托，仅验收放球阶段 1/4。合法发球在击球环节中断，未发生后续合法越网入区。"
+  },
+  "unmet": {
+   "serve_in_target": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 4.72,
+    "log_step": 101,
+    "evidence_zh": "推断：视频 4.68–4.72 s 的挥拍与 log step 101 的服务裁判“先触羽毛、未先触球托”相互对应，说明击球瞬间球拍与羽毛球的相对位姿未让球托成为首次接触部位。这里已发生击打接触，错误是击中部位；仅凭远景与该裁判结果，不能确认挥拍究竟过早还是过晚。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 4.72,
+    "log_step": 101,
+    "source_zh": "日志",
+    "evidence_zh": "推断：log step 68 定义的脚本在检测左手松球后按延迟启动预设八步挥拍序列；log step 86–87 保留这一发球控制逻辑，log step 101 最后一次执行使用延迟参数 0.226 s，服务裁判仍确认先触羽毛。目标及放球后挥拍的顺序正确，但这次放球与挥拍的执行未实现球托先接触，支持归于控制；没有证据将其细分为抓取设计、轨迹跟踪或参数错误。"
+   }
+  ],
+  "gap_zh": "观察：inputs/ 中有本次视频、完整运行日志和成功参考视频，没有本次状态日志。视频为 640×360、25 fps 的场地远景，羽毛球及手指很小，不能直接分辨球托与羽毛的首次接触、毫秒级接触持续、瞬时速度；先触羽毛依据最后一次实时尝试的服务裁判记录。log step 104 中 agent 将球拍滑转、张手推偏羽毛球列为猜测，现有帧和日志未证实，故不据此细分控制根因；DAG 五个原生指标均未由本次评分器输出，记为 null。"
+ },
+ "summary_zh": "本次评分对应最后的第六次发球：机器人站立调整持球及持拍姿态，放球后挥拍，但合法发球入区节点始终未完成，最终为 0/1。视频 4.68–4.72 s 的挥拍对应 log step 101 的服务裁判失败：球拍先触羽毛而非球托，仅完成原生放球阶段 1/4，归为击球位姿不准及控制问题。远景和缺少状态日志使球拍是否滑转、放球是否被手指推偏及具体时序误差无法区分。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点未达成、断点 serve_in_target 在击球环节、时钟（控制步 473≈4.73 s，release 4.41 s、挥拍 4.64 s）与服务裁判记录一致；现象位姿不准与根因控制有日志依据，未发现需改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:58:26Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--cshxrw7/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--cshxrw7/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "70d9a0f66e14231fd07bdf0ae2c0ff2416f6f512b6f1d0bb94bde8bd5a12b354",
+   "video_sha256": "ed985af1cc064ce689ef4b809aade8ef86fa04a28249a1094ca235294193c172",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--lb6jxal.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--lb6jxal.json
new file mode 100644
index 0000000000000000000000000000000000000000..f0fd2b2a74483171cf038a0d6eff94ddb34ff849
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--lb6jxal.json
@@ -0,0 +1,231 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-standard--lb6jxal",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 132,
+   "sim_steps": 7146,
+   "calls": 106,
+   "rows": 7146,
+   "resets": 50,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 6931.4,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.215,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.42,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-standard--lb6jxal/replay.mp4",
+  "duration_s": 1.36,
+  "fps": 25.0,
+  "frames": 34,
+  "sha256": "f2be0da22fc01e405a88986828a675cb5cbbc2425cadc7caa33d18d0eaeb9304"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0,
+      "log_step": 867,
+      "text_zh": "观察：视频 0.00–1.20 s，机器人站在近侧发球区，双臂由初始持球、持拍姿态移至准备姿态；视频 1.24–1.32 s 手臂开始变化，但没有羽毛球向远侧飞行、越网并落入接发区的过程。log step 867 的最终脚本先调整双臂，再打开左手、挥动右臂。"
+     },
+     {
+      "t": 1.32,
+      "log_step": 869,
+      "text_zh": "观察：log step 869 是最后一次实时执行，reset 后为 episode 51，服务返回 executed_rows=132、stages_reached=1 of 4、reason=failed，裁判原文为“the shuttle touched you after you let it go”（放球后羽毛球碰到机器人）。该裁判反馈由机器人服务返回，不是 agent 的完成自述；log step 870 确认同一 episode 已结束且 control_steps=132。"
+     },
+     {
+      "t": 1.32,
+      "log_step": 870,
+      "text_zh": "推断：log step 7 的 control_dt=0.01 s，最终 episode 的 132 个控制步对应视频约 1.32 s；已与视频 1.28–1.32 s 的末段手臂动作及终止画面核对，视频共 1.36 s（末帧为视频 1.32 s）。原生阶段仅确认放球且未出现合法越网落地，故完整发球事件从未达到。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.00–1.20 s，左手持羽毛球、右手持球拍，双臂共同调整至发球准备姿态；随后左手张开放球、右臂启动挥拍（log step 867、869）。此条记录准备阶段两手各持一物的配合，不据此认定球拍击中了球。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–1.32 s，机器人双脚支撑球场地面并保持站立，未见跌倒；远景不用于验证厘米级脚步阈值。"
+     },
+     {
+      "who": "未知",
+      "how": "其他",
+      "note_zh": "观察：log step 869 的本次裁判确认放球后羽毛球再次碰到机器人。候选：碰撞；具体身体接触部位及接触方式在视频 1.24–1.32 s 的远景中无法辨认，不能归为抓握、推/扫或压/擦。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": null
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "serve_in_target",
+   "t": 1.32,
+   "log_step": 869,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 1.24–1.32 s 机器人仍在原地进行放球、挥拍动作，未将球送至远侧；log step 869 的最终 episode 51 在执行 132 行后由裁判以“the shuttle touched you after you let it go”终止，仅确认 1/4 原生阶段。推断：断点是正在执行的合法发球入区，导航锚点取视频约 1.32 s，而不是前面重置测试中的失误。"
+  },
+  "unmet": {
+   "serve_in_target": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "被挡/碰撞",
+    "t": 1.32,
+    "log_step": 869,
+    "evidence_zh": "观察：log step 869 中机器人服务的最终裁判明确报告，羽毛球已被放开，随后碰到机器人，立即判负；视频 1.24–1.32 s 球仍处于机器人手臂附近，未出现远侧飞行。推断：此次放球、挥拍中的球与机器人发生了非预期碰撞；不把这个事件另记为违反约束，也不声称看清了具体指节。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "source_zh": "日志",
+    "evidence_zh": "观察：log step 867 的最终控制脚本在左手打开后仍以 A3[:,3:10]=q_left 保持左臂姿态，右臂按固定延迟和插值挥拍；log step 869 的同次执行反馈是放球后碰身、只确认放球阶段。推断：这套双臂放球与挥拍控制未实现羽毛球脱离身体的必要间隙，导致发球中断。没有本次逐步接触与跟踪数据，不能进一步确定是具体接触几何、实际跟踪偏差还是时序参数问题。"
+   }
+  ],
+  "gap_zh": "本目录没有状态日志；视频只有远景，已逐帧检查视频 0.00–1.32 s，尤其视频 1.00–1.32 s 的放大图，仍不能分辨羽毛球再接触的机器人部位、球拍是否短暂触球及毫秒级接触顺序。最终服务裁判足以确认放球后碰身失败，但不报告五项原生指标，故这些值均为 null；根因仅归至控制，不指定子类。log step 876 自述已击球且避免干涉，与 log step 869 的最终裁判结果不符，不能作为本次合法击球的证据。"
+ },
+ "summary_zh": "最后一次实时尝试（episode 51）中，机器人完成持球、持拍准备并放球，但在视频约 1.32 s 因羽毛球放开后碰到机器人而被裁判终止，合法发球入区未达到，最终为 0/1。最终脚本的双臂放球、挥拍控制未保证球与身体分离；远景及缺少状态日志使具体碰撞部位、瞬时球拍接触和控制子类无法核实。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "最终 episode 51 的视频（1.36 s，132 控制步，dt=0.01）与 log step 867/869/870 一致：放球后羽毛球碰到机器人，服务裁判终止，仅 1/4 阶段；serve_in_target 未达成、断点、接触与现象/根因均有证据支持，未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:55:25Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--lb6jxal/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--lb6jxal/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "4768ab70d065edf7e127ee959a29ea937c92116a4a6a3fa53371f73979dbe313",
+   "video_sha256": "f2be0da22fc01e405a88986828a675cb5cbbc2425cadc7caa33d18d0eaeb9304",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--mvet3aa.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--mvet3aa.json
new file mode 100644
index 0000000000000000000000000000000000000000..854cb8631b8c602177928fcdd77eb52ab0171ccc
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--mvet3aa.json
@@ -0,0 +1,221 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-standard--mvet3aa",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 783,
+   "sim_steps": 24174,
+   "calls": 98,
+   "rows": 24174,
+   "resets": 42,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 7028.4,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.155,
+   "peak_processes": 11,
+   "peak_disk_gb": 0.887,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-standard--mvet3aa/replay.mp4",
+  "duration_s": 7.84,
+  "fps": 25.0,
+  "frames": 196,
+  "sha256": "1071b66a3e94c2c00d523b6e86ca372ae2afbd4a8691301c7fedaae2f33dd86b"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–6.80 s，机器人在近侧发球区双脚站立，左手持白色羽毛球调整放球姿态，右手握球拍逐渐收拍准备；尚无球飞向远侧接发球区。"
+     },
+     {
+      "t": 7.76,
+      "log_step": 366,
+      "text_zh": "观察：视频 6.80–7.80 s 已逐帧检查，机器人保持站立，右手球拍在视频约 7.76–7.80 s 快速挥动，未见羽毛球形成越网飞向远场的轨迹。log step 366 是最后一次重置后的实时提交，服务返回 executed_rows=783、done=True、outcome=\"the racket's forward movement stopped before the hit\"；reward.n_actions 同为 783，graded_by_live=1。该服务结果明确为击球前前挥停止，合法发球入区未完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.00–7.60 s，左手包握羽毛球并调整放球姿态，右手握住球拍拍柄同时准备挥拍；随后左手放球、右手挥拍。log step 366 的服务结果未记录到击球，因此不将球拍与羽毛球的击打作为已发生接触；这里记录实际双手持物配合。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–7.80 s，机器人双脚在球场地面支撑身体，调整双臂及末段挥拍期间仍保持站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": null
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "serve_in_target",
+   "t": 7.76,
+   "log_step": 366,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 7.76–7.80 s，机器人正在持拍挥动，未出现球飞向对角接发球区的后续过程；最后一次实时发球提交在 log step 366 返回“the racket's forward movement stopped before the hit”，done=True。推断：此次已启动发球操作，但前挥停止时尚未击中羽毛球，断点为合法发球入区。"
+  },
+  "unmet": {
+   "serve_in_target": false
+  },
+  "phenomena": [
+   {
+    "cat": "接触",
+    "item": "没建立接触",
+    "t": 7.76,
+    "log_step": 366,
+    "evidence_zh": "观察：视频 7.48–7.80 s 的近身裁剪中可见右手球拍末段挥动，没有羽毛球被击出并越网的飞行；log step 366 的本次服务终止反馈明确为前挥在击球前停止。推断：本次发球所需的球拍—羽毛球击打接触尚未建立；不将同一失误另标为动态时机，因为现有证据不能区分时间偏差与空间偏差。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "观察：log step 366 中 agent 的脚本修改左肩横向姿态并将 I_REL 从 11 改为 17，重建固定关节动作序列后重置、以 dt=0.01 提交执行；本次服务仍返回前挥在击球前停止。视频约 7.76–7.80 s 确有挥拍而无击出飞行。推断：放球与挥拍的运动配合未使球拍在有效前挥期间接到球，归为控制；没有本次精细轨迹与接触测量，不进一步指定参数调试、逆解或反馈控制。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "无状态日志；远景中羽毛球及手指很小，无法仅凭帧图确定精确脱手时刻、拍球最近距离及毫秒级接触。未击球判断由本次最后提交的服务反馈与视频共同支持；log step 365 的几何测算属于前一次 v18 调试，不能当成本次 v19 的实际轨迹，因此具体空间偏差、时间偏差及控制子类仍不确定。评分器未报告 DAG 所列五项原生指标，均填 null。"
+ },
+ "summary_zh": "本次评分为 log step 366 重置后最后一次实时发球：机器人完成持球调整并挥拍，但合法发球入区节点始终未达到，最终 0/1。视频末段未见球被击出越网，服务明确反馈前挥在击球前停止；失败表现为没建立击打接触，根因保留为控制配合失败。无状态日志且球体很小，无法确定具体失配方向或精确释放时刻。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "标签与视频、log step 366 的服务反馈一致：合法发球入区未达到，断点、没建立接触、控制根因均有证据支持，未发现需改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:53:26Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--mvet3aa/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--mvet3aa/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "eae716d05a914dc69c7079299310c13d1704b53de1e963b337cde1f00cbb9c8f",
+   "video_sha256": "1071b66a3e94c2c00d523b6e86ca372ae2afbd4a8691301c7fedaae2f33dd86b",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--sahqtl6.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--sahqtl6.json
new file mode 100644
index 0000000000000000000000000000000000000000..bbd21373f0e1585ec62b5a8ca01f2e0d3fbf03db
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--sahqtl6.json
@@ -0,0 +1,196 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-standard--sahqtl6",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 1,
+   "sim_steps": 1,
+   "calls": 31,
+   "rows": 1,
+   "resets": 0,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 1795.3,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 4,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-standard--sahqtl6/replay.mp4",
+  "duration_s": 0.04,
+  "fps": 25.0,
+  "frames": 1,
+  "sha256": "7f28443388c68edd77037540350bf291bc019550e5b1cbe4705cd6e929f008a5"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：已查看本次完整视频，视频仅有 0.00 s 一帧（总长 0.04 s）；机器人直立于近侧球场并在右手处持拍，画面未显示发球、羽毛球越网或落入远侧对角区。"
+     },
+     {
+      "t": null,
+      "log_step": 57,
+      "text_zh": "观察：log step 57 的脚本只向实时机器人服务提交一行初始 stand 指令，用于读取站姿力矩；服务返回仿真 0.01 s、done=False、holding_left=1.0、holding_right=1.0。此后日志只有本地逆运动学等计算，没有继续提交发球控制；评分器记录 n_actions=1、sim_steps=1、resets=0、live_success=0，未形成合法发球入区事件。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": null
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "unknown",
+   "t": null,
+   "log_step": 79,
+   "where": "首步前",
+   "evidence_zh": "观察：log step 79 明确返回 Gemini 模型过滤拦截消息；此前唯一实时控制在 log step 57 只维持初始站姿，服务返回 done=False，随后未提交其他控制。视频 0.00 s 唯一帧也未显示发球动作。推断：评分尝试尚未结束或形成已确定的物理失败，便被提供方过滤中断；依 CASES.md FAIL-5 与 DECISIONS.md D37，按外因中断处理，不能强选机器人失败步骤。"
+  },
+  "unmet": {
+   "serve_in_target": false
+  },
+  "phenomena": [],
+  "root_causes": [],
+  "gap_zh": "观察：inputs/ 中有本次单帧视频与完整会话日志，无状态日志；log step 57 服务仍返回 done=False，log step 79 是提供方过滤拦截，没有已完成发球的交付或裁判事件记录。推断：这是结果确定前的外因中断，应重试并排除出机器人失败分析；不能将未启动发球标成“没尝试”，也没有证据支持任何机器人根因。单帧无法核实瞬时接触过程，评分器未报告击球高度、速度、角度或落点指标。"
+ },
+ "summary_zh": "本次记录的合法发球入区节点未达成，终态进度为 0/1；视频 0.00 s 唯一帧仅显示机器人持拍直立，log step 57 只执行了一行初始站姿诊断，服务仍返回 done=False。log step 79 随后发生 Gemini 提供方过滤拦截，属于评分结果确定前的外因中断，断点记 unknown／首步前，现象与根因留空并应重试。视频仅一帧且无状态日志，不能核实精细接触或原生击球、落点指标；本地仿真与逆解计算不计为本次节点进度。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "标注与证据一致：视频仅一帧，节点未达成；整个会话只提交了一次 10 ms stand 指令（sim_total.submit=1），log step 79 为 Gemini 过滤拦截，发球结果未定，按外因中断（FAIL-5）处理，断点 unknown/首步前、两层为空且有 gap_zh，合理。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:56:41Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--sahqtl6/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--sahqtl6/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "c7c89b3fbc262a9b97df65c064e5a24a7b14a5e228c9813c6437fe6c3906def5",
+   "video_sha256": "7f28443388c68edd77037540350bf291bc019550e5b1cbe4705cd6e929f008a5",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--xsd7hrj.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--xsd7hrj.json
new file mode 100644
index 0000000000000000000000000000000000000000..a8b85852e336b904dc80d0d806985df1431ee30b
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--xsd7hrj.json
@@ -0,0 +1,214 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-standard--xsd7hrj",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "live_success": 1,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 494,
+   "sim_steps": 494,
+   "calls": 13,
+   "rows": 17,
+   "resets": 0,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 407.1,
+   "video_rendered": 1,
+   "first_success": 1,
+   "first_reason_code": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.311,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.868,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-standard--xsd7hrj/replay.mp4",
+  "duration_s": 4.96,
+  "fps": 25.0,
+  "frames": 124,
+  "sha256": "fbc297911e7d048129e60ec52deba503f1a12a8ec8b34ee7a4cd45810d0e2fd0"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.94,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–3.48 s，机器人站在近侧发球区，持球、持拍调整双臂姿态，尚未出现球飞向对场的发球结果；双脚持续在场地上支撑。"
+     },
+     {
+      "t": 3.72,
+      "log_step": 35,
+      "text_zh": "观察：逐帧检查视频 3.40–3.96 s，白色羽毛球先由左手附近下落，机器人在视频约 3.72 s 右臂挥拍，视频 3.76 s 白球已离开手拍附近并升高，随后机器人维持随挥姿态；log step 35 的实时服务记录执行击球指令后 control_steps=371、episode=1，log step 37 为 control_steps=373，右手持物传感值仍为 1。"
+     },
+     {
+      "t": 4.04,
+      "log_step": null,
+      "text_zh": "观察：视频 3.96–4.12 s 可见微小白球继续向网后飞行，视频 4.04 s 白球投影在网顶上方；已逐帧检查视频 3.96–4.92 s，后段白球难以与背景分辨，末帧未清楚显示首次触地接触点。"
+     },
+     {
+      "t": 4.92,
+      "log_step": 40,
+      "text_zh": "观察：log step 40 中，机器人实时服务（含本次发球的程序裁判）返回 episode=1、control_steps=494、sim_time_s=4.94、done=true、reason=success、outcome=success、stages_reached=4、stages_total=4；inputs/run.json 的 reward 同时记录 success=1、graded_by_live=1、resets=0。该结果来自本次唯一实时评分局，不是 agent 自述或另一次测试。"
+     },
+     {
+      "t": 4.94,
+      "log_step": 40,
+      "text_zh": "推断：log step 7 的服务规格给出 control_dt=0.01 s，以本次评分局控制步数映射视频时钟；log step 35–37 的 control_steps=371–373 对应视频 3.71–3.73 s，已与视频约 3.72 s 的挥拍核对，494 个控制步也与视频总长 4.96 s 相符。因此将实时裁判成功时刻 494×0.01=4.94 s 标为视频约 4.94 s 达成“合法发球入区”，之后事件锁存；该时刻比最后采样帧视频 4.92 s 晚约 0.02 s，合法落地及精细击球条件依据裁判验收，未声称画面直接拍到触地。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "候选：击打。视频约 3.72–3.76 s 机器人右手所持球拍挥向下落羽毛球，随后球上升离开；结合 log step 40 的四阶段合法发球成功，判定球拍实际击中羽毛球。左手此前持球并先行松开；右手握拍链路不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 0.00–4.92 s 机器人双脚支撑球场地面，调整双臂、挥拍和随挥时均未见移步或倒地。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": null
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": null,
+ "summary_zh": "观察：机器人完成左手放球、右手挥拍及羽毛球向网后飞行，log step 40 的本次实时裁判确认四阶段全部成功，唯一节点“合法发球入区”在视频约 4.94 s 达成，最终 1/1。观察：本次与参考视频均为远景，无状态日志；首次触地、球托初触及毫秒级接触条件无法由画面独立核实，故以本次裁判验收补足，五项原生指标未报告数值，均保留 null。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "单节点成功标注与证据一致：log step 40 的实时裁判 done=true、success、4/4、sim_time_s=4.94、control_steps=494，时钟映射(0.01 s/步)与 step 35/37 的 371/373 步对应视频约3.72 s 核对无误；contact_actual 合理，无失败层，未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:01:05Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--xsd7hrj/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--xsd7hrj/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "6aa96cc29d0cb45c9beb63f3a3a82ab8f8f64c2e1424186fb953aa7d2944ca6e",
+   "video_sha256": "fbc297911e7d048129e60ec52deba503f1a12a8ec8b34ee7a4cd45810d0e2fd0",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--zyb6fty.json b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--zyb6fty.json
new file mode 100644
index 0000000000000000000000000000000000000000..c3591d297aa02966ed1012d88dd1bb9d398e01f5
--- /dev/null
+++ b/phase2/embodied-olympics-fun/badminton_serve/trajectories/embodied-olympics-fun-badminton-serve-i00-standard--zyb6fty.json
@@ -0,0 +1,226 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-badminton-serve-i00-standard--zyb6fty",
+ "taskKey": "embodied-olympics-fun/badminton_serve",
+ "dag_sha256": "e5772140aff5f054a271f90fd295419869c755e083b6526601e08a99bee573e1",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 218,
+   "sim_steps": 8015,
+   "calls": 350,
+   "rows": 7663,
+   "resets": 37,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 7060.3,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.351,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.955,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-badminton-serve-i00-standard--zyb6fty/replay.mp4",
+  "duration_s": 2.2,
+  "fps": 25.0,
+  "frames": 55,
+  "sha256": "520ba939beaa65327f4820bb6aef5fc8ff32fc4d8db4aacdb54e8617dd267236"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "serve_in_target": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–2.16 s 的远景中，机器人双脚支撑在近侧发球区，先调整双臂和球拍，末段右手持拍向左手附近挥动；未见羽毛球飞越球网进入远侧目标区。成功参考视频 1.72–2.96 s 可见球离开机器人向远侧飞行，参考只用于识别该视觉线索。"
+     },
+     {
+      "t": 2.16,
+      "log_step": 460,
+      "text_zh": "观察：log step 459 启动最后一次 reset 发球，log step 460 返回 episode 38，末次 release/stroke 的裁判 outcome 为“the racket hit the feathers first, not the base of the shuttle”，stages=1、done=true；这是机器人服务返回的裁判判定，非 agent 自述。视频 2.12–2.16 s 可见末段挥拍，但分辨率不足以辨别羽毛与球托的首次接触。"
+     },
+     {
+      "t": 2.16,
+      "log_step": 461,
+      "text_zh": "推断：log step 461 的末次 episode 38 状态为 failed、control_steps=218，与评分 reward.n_actions=218 相同；服务 control_dt=0.01 s（log step 25），218 个控制步对应仿真 2.18 s，与视频 2.16 s 末帧及 2.20 s 总长吻合，并已核对视频 2.12–2.16 s 的挥拍末段。故本次评分局为最后一次发球，已放球但首次击球不合法，合法发球入区节点从未成立；会话中早先 reset 的结果不计入。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 0.00–2.16 s 机器人右手持球拍调整并挥动；log step 460 的末次裁判确认球拍实际击中羽毛球羽毛。候选：击打；现有方式无击打，不能以推/扫或压/擦代替。左手此前持球后张开放球，属先后动作；右手握拍链路不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–2.16 s 机器人双脚持续支撑场地，发球调整和挥拍期间未见跌倒或明显移步；不据远景判断毫米级脚步阈值。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": 1.15,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "shuttle_angle_deg",
+   "unit": "deg",
+   "threshold": 0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "landing_x_m",
+   "unit": "m",
+   "threshold": [
+    4.43,
+    9.15
+   ],
+   "direction": "target",
+   "value": null
+  },
+  {
+   "name": "landing_y_m",
+   "unit": "m",
+   "threshold": [
+    0.43,
+    3.04
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "serve_in_target",
+   "t": 2.16,
+   "log_step": 460,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 2.08–2.16 s 机器人右臂挥拍靠近放球位置，未出现随后越网飞行；最后一次评分局的服务裁判在 log step 460 返回“先击中羽毛而非球托”、stages=1、done=true，log step 461 确认 episode 38 已失败。断在正在执行的合法发球入区步骤，视频 2.16 s 为末段击球的近似导航锚点。"
+  },
+  "unmet": {
+   "serve_in_target": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 2.16,
+    "log_step": 460,
+    "evidence_zh": "观察：视频 2.08–2.16 s 可见机器人挥拍至左手附近，log step 460 的本次裁判明确判定球拍先接触羽毛、没有球托先触拍。推断：球拍与羽毛球的相对击球位置或朝向未对准要求的球托，属于击球部位偏差；没有证据进一步断言是挥拍过早或过晚。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：agent 在 log step 457 自述要恢复放球对齐并小幅调整手腕以提高击球方向，随后 log step 459–460 实际执行发球，裁判仍检出羽毛先触拍；视频 2.08–2.16 s 也显示机器人已进入挥拍动作。目标与放球后击球的顺序明确，但实际挥拍未实现球托先接触的控制目标，因此归为控制；现有证据不足以区分接触设计、轨迹跟踪或具体参数问题。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "本次及成功参考均无状态日志；视频远景只有 640×360、25 fps，不能直接分辨毫秒级球托/羽毛初触或重建接触点，首次接触部位依据已核对为最后评分局的服务裁判返回。缺少最终控制脚本完整修改内容与球拍、羽毛球接触轨迹，控制根因不能细分；DAG 的五项原生指标在评分 reward 中均未报告，value 均为 null。"
+ },
+ "summary_zh": "本次评分为最后的 episode 38：机器人完成发球准备并放球挥拍，但合法发球入区节点未达到，最终进度 0/1。视频 2.08–2.16 s 的末段挥拍与 log step 460 的裁判结果对应，球拍先击中羽毛而非球托，断点为操作中的合法发球入区，归为击球位姿不准及控制问题。远景无法分辨精细接触，控制子类与原生指标缺乏证据；早先 reset 的测试进展不计入本次评分。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "评分局为最后的 episode 38（control_steps=218，约 2.18 s，与视频 2.2 s 吻合）；服务裁判返回球拍先击中羽毛，serve_in_target 未达成，断点、位姿不准与控制根因均有证据支持，与先例一致，未发现需改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:01:43Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--zyb6fty/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-badminton-serve-i00-standard--zyb6fty/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "523e1c0c1167c57668c0e29c65f3f980354e2437ce3b1a34f86cb9411357540d",
+   "video_sha256": "520ba939beaa65327f4820bb6aef5fc8ff32fc4d8db4aacdb54e8617dd267236",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--ebemxbr.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--ebemxbr.json
new file mode 100644
index 0000000000000000000000000000000000000000..4eb7fa2160e75057457302551b505e50dd80eaa1
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--ebemxbr.json
@@ -0,0 +1,222 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-privileged--ebemxbr",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 224,
+   "sim_steps": 224,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 3.1,
+   "bad_trajectory": 0,
+   "metric_contact_impulse_n_s": 20.64,
+   "metric_contact_s": 0.049,
+   "metric_hit_height_m": 1.419,
+   "metric_impulse_n_s": 16.06,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 44,
+   "agent_resets": 7439,
+   "agent_replays": 6,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.188,
+   "peak_processes": 13,
+   "peak_disk_gb": 0.846,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-privileged--ebemxbr/replay.mp4",
+  "duration_s": 2.28,
+  "fps": 25.0,
+  "frames": 57,
+  "sha256": "78fb7ce7961dfdc5cd0d2f6acf06541f8b0dedc21fd759ae4bcf7788bf539368"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 2,
+  "final": 2,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [
+     [
+      0.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.12,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.08 s 机器人右臂抬起向沙袋伸出；视频 0.12–0.20 s 右拳末端抵达红色沙袋白色环带，沙袋开始向远离机器人的方向摆动。视频 0.24–0.28 s 右拳明显收回并与沙袋分离，未见左手击袋。"
+     },
+     {
+      "t": 0.24,
+      "log_step": 216,
+      "text_zh": "观察：log step 216 的最终交付验证结果记录两个新进程重放均成功、224 个动作行完全相同、终态哈希一致；裁判在仿真 0.213 s 记录 strike，impulse_n_s=16.06、contact_s=0.049、hit_height_m=1.419，与 inputs/run.json 的最终评分字段一致。"
+     },
+     {
+      "t": 0.24,
+      "log_step": 216,
+      "text_zh": "推断：结合视频 0.12–0.28 s 的一次触袋、摆离和收拳，以及最终交付裁判的有效击打记录，击中沙袋节点已完成；以视频 0.24 s 的分离帧作近似完成锚点。精确冲量及拳峰接触区域依据裁判验收，不能由画面独立测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "观察：视频 0.12–0.20 s 机器人右手末端拳垫击中沙袋白带，随后迅速撤回；候选接触名“击打”，现有接触方式没有短促冲击的对应值，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–0.28 s 机器人双脚接地，腿和躯干支承出拳，身体保持站立。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.28,
+      "log_step": null,
+      "text_zh": "观察：视频 0.24–0.32 s 机器人右肘弯曲，右拳从沙袋前收向右肩附近；随后至视频 2.24 s，双脚持续接地，躯干直立、位置基本不变，右拳保持在肩前附近，沙袋摆回时未见再次接触。已逐帧检查视频 1.28–2.24 s 的末段。"
+     },
+     {
+      "t": 2.24,
+      "log_step": 216,
+      "text_zh": "观察：log step 216 的最终交付及两次独立重放均在仿真 2.24 s、控制步 224 记录 guard 和 success，failed=null；inputs/run.json 的评分结果为 success=1、replay_success=1、deterministic=1、stages_reached=2、fell=0。视频 2.24 s 末帧仍呈收拳站立姿态。"
+     },
+     {
+      "t": 2.24,
+      "log_step": 216,
+      "text_zh": "推断：视频中的收拳站立与最终交付的 guard 验收一致，因此护架连续保持达标，节点在视频末帧 2.24 s 完成；不把视频约 0.28 s 开始呈护架姿态误记为已完成保持。无状态日志，右掌至肩距离、躯干余弦及骨盆速度的精确阈值由裁判成功结果支持。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.28–2.24 s 机器人以双脚支撑地面，腿与躯干保持站立；右拳收回肩前，未再接触沙袋。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": 16.06
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": 0.049
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": 1.419
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人完成有效击袋和护架站稳两步：视频约 0.24 s 已击中白带并收拳分离，随后保持站立，视频 2.24 s 完成护架验收，最终 2/2 节点成立。评分器记录有效冲量 16.06 N·s、接触 0.049 s、首次接触高度 1.419 m，最终交付两次独立重放均成功且终态一致（log step 216）。没有状态日志，精确拳峰接触位置及护架距离、速度阈值不能从录像独立测量，依据最终裁判验收确认。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "两节点均成立：视频约 0.12–0.28 s 右拳击中白带并收回，final 在视频 2.24 s（控制步 224）完成，与 log step 216 的裁判事件（strike 仿真 0.213 s、guard/success 2.24 s）及 run.json 评分字段一致；时间钟明确，contact_actual 合理，无失败故 failure=null。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:46:19Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--ebemxbr/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--ebemxbr/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "26a97cfd052c7111c362c52a56b278fc9e31f8402aec49cd6d9b5b3c22b25fd9",
+   "video_sha256": "78fb7ce7961dfdc5cd0d2f6acf06541f8b0dedc21fd759ae4bcf7788bf539368",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--gvnh9sq.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--gvnh9sq.json
new file mode 100644
index 0000000000000000000000000000000000000000..7bfafb77fe38916ece9d12e17edc25549665daa6
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--gvnh9sq.json
@@ -0,0 +1,222 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-privileged--gvnh9sq",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 1490,
+   "sim_steps": 425,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 3.11,
+   "bad_trajectory": 0,
+   "metric_contact_impulse_n_s": 25.86,
+   "metric_contact_s": 0.046,
+   "metric_hit_height_m": 1.358,
+   "metric_impulse_n_s": 21.16,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 122,
+   "agent_resets": 1130,
+   "agent_replays": 63,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.507,
+   "peak_processes": 11,
+   "peak_disk_gb": 1.049,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-privileged--gvnh9sq/replay.mp4",
+  "duration_s": 4.28,
+  "fps": 25.0,
+  "frames": 107,
+  "sha256": "d86b330b0a2868c461729b12f0ce8de5201c5fe9d45239085e081a9cd3ad8fac"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 2,
+  "final": 2,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [
+     [
+      0.96,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.8,
+      "log_step": null,
+      "text_zh": "观察：视频 0.56–0.84 s，机器人右臂由屈肘蓄势转为向前伸出，右手末端打到红色沙袋白带；左手留在身前，双脚支承身体，沙袋随后向远离机器人的方向摆动。"
+     },
+     {
+      "t": 0.96,
+      "log_step": null,
+      "text_zh": "观察：视频 0.96–1.04 s，右手末端与白带之间已出现可见间隙，沙袋继续摆离。inputs/run.json 的评分器字段报告 metric_impulse_n_s=21.16、metric_contact_s=0.046、metric_hit_height_m=1.358，stages_reached=2、success=1；冲量和接触时长均满足击打验收。"
+     },
+     {
+      "t": 0.96,
+      "log_step": 264,
+      "text_zh": "推断：结合视频中的一次击中、分离和评分器通过结果，击中沙袋节点已达成；以视频 0.96 s 的清晰分离帧作完成锚点，实际结算时刻有约 0.2 s 不确定性。log step 264 对最终交付文件重新生成并两次重放，输出均含 strike、guard 成功，指标与本次评分一致；未将早期试验中的击打计入。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "观察：视频 0.72–0.84 s，机器人右手末端拳垫击中沙袋白色环带。候选接触名为“击打”；现有方式中无短促冲击条目，不属于抓握、推/扫或压/擦；具体拳峰末端接触区由评分器成功结果支持。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.56–1.04 s，机器人以左右脚接地支承出拳，腿和躯干维持直立，脚有原地调整。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.32,
+      "log_step": null,
+      "text_zh": "观察：视频 1.04–2.32 s，机器人右肘逐渐弯曲、右手从袋前收回身前；视频 2.28–4.24 s，右臂保持收回姿态，躯干直立，双脚原地小幅交替调整，未见手、膝或躯干落地，右手与沙袋保持分离。"
+     },
+     {
+      "t": 4.24,
+      "log_step": 267,
+      "text_zh": "观察：视频 3.24–4.24 s 的全部逐帧画面中，机器人继续保持屈肘站立至最后一帧。inputs/run.json 评分器报告 success=1、stages_reached=2/2、fell=0；log step 267 对最终交付文件的重放也输出 success，指标与评分记录一致。"
+     },
+     {
+      "t": 4.24,
+      "log_step": null,
+      "text_zh": "推断：评分器两阶段全部通过，且片尾姿态与护架站稳相符，因此连续保持验收完成、节点终值为真；将视频 4.24 s 最后一帧作为约略完成锚点，而非把开始收拳或刚进入护架的时刻算作完成。无状态日志，不能仅从画面独立量出右掌至肩距离、骨盆速度及连续 200 控制步。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 1.04–4.24 s，机器人收回右臂后以左右脚接地、原地小幅调整支承身体；右手不再触袋，腿和躯干完成站立平衡保持。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": 21.16
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": 0.046
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": 1.358
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人完成一次右拳击白带并与沙袋分离，随后收拳护架站立至视频 4.24 s，两个节点最终均完成。评分器记录有效冲量 21.16 N·s、接触 0.046 s、命中高度 1.358 m，且两阶段通过；最终交付文件的重放核验与之相符。无状态日志，精确拳峰接触位置、连续护架阈值依赖评分器结果，节点时刻按视频帧近似定位；成功参考视频仅用于比较姿态线索。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "两个节点均成功，与评分器（success=1、stages 2/2、fell=0、冲量 21.16、接触 0.046 s）及视频帧一致；时刻按视频时钟，击打 0.96 s、护架 4.24 s（sim_steps 425≈4.25 s）合理，接触标注符合 S16，无失败层。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:45:09Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--gvnh9sq/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--gvnh9sq/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "f94272e78c17ad40a7dc8d2e7409fdf2381ea415913ddba718ac917e6245ec08",
+   "video_sha256": "d86b330b0a2868c461729b12f0ce8de5201c5fe9d45239085e081a9cd3ad8fac",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--iihc6lf.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--iihc6lf.json
new file mode 100644
index 0000000000000000000000000000000000000000..159a5bec57fbae33eb033e94c75d4a620b90193a
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--iihc6lf.json
@@ -0,0 +1,222 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-privileged--iihc6lf",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 373,
+   "sim_steps": 373,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 3.66,
+   "bad_trajectory": 0,
+   "metric_contact_impulse_n_s": 25.51,
+   "metric_contact_s": 0.028,
+   "metric_hit_height_m": 1.401,
+   "metric_impulse_n_s": 21.03,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 10,
+   "agent_resets": 0,
+   "agent_replays": 3,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.706,
+   "peak_processes": 7,
+   "peak_disk_gb": 0.246,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-privileged--iihc6lf/replay.mp4",
+  "duration_s": 3.76,
+  "fps": 25.0,
+  "frames": 94,
+  "sha256": "f3912a413e02bdaa0a5caa489bd654b7cc4b0dd3414e45b8c2fc986b52d4fc43"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 2,
+  "final": 2,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [
+     [
+      0.68,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.6,
+      "log_step": null,
+      "text_zh": "观察：视频 0.48–0.60 s，机器人右臂快速前伸，右拳末端到达红色沙袋的白色环带；视频 0.64–0.72 s，右肘弯曲收拳、拳袋分离，沙袋随后明显摆离。双脚在地面支撑，机器人未倒地。"
+     },
+     {
+      "t": 0.68,
+      "log_step": null,
+      "text_zh": "观察：inputs/run.json 的本次评分器（仿真裁判）记录 success=1、replay_success=1、stages_reached=2/2，metric_impulse_n_s=21.03 N·s、metric_contact_s=0.028 s、metric_hit_height_m=1.401 m，确认该次击打有效。"
+     },
+     {
+      "t": 0.68,
+      "log_step": null,
+      "text_zh": "推断：结合视频 0.60–0.72 s 的击中与分离过程及本次评分器通过结果，将有效击打结算完成粗定位于视频 0.68 s；25 fps 画面不能独立测出毫秒级接触起止或拳峰接触深度。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "观察：视频 0.56–0.60 s，机器人右手末端击中沙袋白带后收回；候选接触名为“击打”，属于短促冲击，现有怎么碰条目无对应值，不是抓握或持续推压。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.40–0.72 s，机器人左右脚接地支承出拳，腿和躯干维持站立。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 1.8,
+      "log_step": null,
+      "text_zh": "观察：视频 0.68–0.84 s，机器人右肘弯曲、右拳从袋前收回至肩前附近；视频 0.80–1.60 s，机器人以脚步调整位置，之后明显趋于稳定。"
+     },
+     {
+      "t": 3.72,
+      "log_step": null,
+      "text_zh": "观察：视频 1.80–3.72 s，机器人保持弯肘护架与直立姿态，脚在地面支撑、位置基本稳定；逐帧检查视频 2.72–3.72 s 未见倒地或再次触袋。inputs/run.json 记录 stages_reached=2/2、progress=1.0、success=1、fell=0。"
+     },
+     {
+      "t": 3.72,
+      "log_step": null,
+      "text_zh": "推断：本次评分器成功结果确认击打后的护架距离、直立程度、骨盆速度及连续 200 个控制步保持条件均已通过；结合末尾视频帧，将护架保持完成定位于视频约 3.72 s，而非收拳初始时刻。无状态日志，不能逐控制步独立核对阈值及计时起点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.80–1.60 s，机器人双腿通过脚步调整吸收击打后的运动，随后双脚接地支撑直立护架至视频 3.72 s；收回的右手与沙袋保持分离。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": 21.03
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": 0.028
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": 1.401
+  }
+ ],
+ "failure": null,
+ "summary_zh": "本次评分轨迹完成击中沙袋与护架站稳两个节点：视频约 0.68 s 已击袋并分离，视频末尾约 3.72 s 完成护架连续保持，评分器确认成功且未跌倒。右拳对白带的短促击打及收拳、双脚调整后站稳均可见，原生指标为有效冲量 21.03 N·s、接触 0.028 s、击中高度 1.401 m。无状态日志，毫秒级接触、拳峰深度及护架连续阈值无法从画面独立测量，精确验收采用本次评分器结果，节点时刻为视频近似锚点。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "两个节点（击中沙袋约视频 0.68 s、护架站稳约视频 3.72 s）与视频和评分器结果一致，接触标注合理，无失败故而 failure 为 null，未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:53:35Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--iihc6lf/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--iihc6lf/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "0b08fcfcdc5f6a90ea5866871fe1d235667c11bdd64b30adb2803a2d59064c03",
+   "video_sha256": "f3912a413e02bdaa0a5caa489bd654b7cc4b0dd3414e45b8c2fc986b52d4fc43",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--mpenplb.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--mpenplb.json
new file mode 100644
index 0000000000000000000000000000000000000000..470af81bfe8d62e3bf48ddbe8bbd1c9e88d8bbe1
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--mpenplb.json
@@ -0,0 +1,222 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-privileged--mpenplb",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 317,
+   "sim_steps": 317,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 3.64,
+   "bad_trajectory": 0,
+   "metric_contact_impulse_n_s": 21.17,
+   "metric_contact_s": 0.056,
+   "metric_hit_height_m": 1.563,
+   "metric_impulse_n_s": 14.88,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 34,
+   "agent_resets": 1607,
+   "agent_replays": 1,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 2.253,
+   "peak_processes": 18,
+   "peak_disk_gb": 0.239,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-privileged--mpenplb/replay.mp4",
+  "duration_s": 3.2,
+  "fps": 25.0,
+  "frames": 80,
+  "sha256": "1c23b1c7ba19324690d7ca3f63177cbcc268a797400d4c3d742e0be3aaf24d13"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 2,
+  "final": 2,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [
+     [
+      1.04,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.96,
+      "log_step": null,
+      "text_zh": "观察：视频 0.80–0.92 s 机器人转动躯干并向沙袋伸出右拳，视频 0.96–1.00 s 右拳末端抵达白色环带上部；视频 1.04–1.20 s 沙袋摆离，右肘弯曲、拳头后撤。双脚始终在地面支撑。"
+     },
+     {
+      "t": 1.04,
+      "log_step": 30,
+      "text_zh": "观察：log step 30 对刚保存的交付文件执行官方重放，裁判返回 strike 事件（仿真 1.028 s）、impulse_n_s=14.88、contact_s=0.056、hit_height_m=1.563，且 failed=null；inputs/run.json 的最终评分器字段报告相同指标，replay_success=1、deterministic=1。"
+     },
+     {
+      "t": 1.04,
+      "log_step": 30,
+      "text_zh": "推断：log step 6 记录控制周期为 0.01 s；log step 30 的交付重放共 317 控制步、仿真时长 3.17 s，与视频 3.20 s 相符，且其击打事件与视频 0.96–1.04 s 的拳袋接触及随后分离相符。因此将有效击打结算近似锚定在视频 1.04 s，精确冲量与接触时长采用裁判记录，不由画面摆幅估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "机器人右手末端拳垫短促击打沙袋白色环带；候选名称“击打”，现有接触方式无短促冲击条目，不是抓握或持续推袋。视频 0.96–1.04 s 可见拳袋相遇，随后分离。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "出拳和躯干转动期间，机器人左右脚接地承重，双腿支撑身体，视频 0.80–1.20 s 未见跌倒。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.16,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 1.2,
+      "log_step": null,
+      "text_zh": "观察：视频 1.12–1.40 s 机器人右肘弯曲，右拳从袋前收至肩部附近，躯干逐渐转回；视频 1.60–3.16 s 右拳保持收回、身体基本直立，双脚留在原处支撑。已逐帧检查视频 2.16–3.16 s，未见跌倒或再次触袋。"
+     },
+     {
+      "t": 3.16,
+      "log_step": 30,
+      "text_zh": "观察：log step 30 对交付文件的官方重放返回 guard 与 success 事件，均在仿真 3.17 s，stages=[strike,guard]、steps=317；最终评分器字段也报告 stages_reached=2/2、success=1、fell=0，与视频末帧的收拳站姿一致。"
+     },
+     {
+      "t": 3.16,
+      "log_step": 30,
+      "text_zh": "推断：按 log step 6 的每控制步 0.01 s 和 log step 30 的交付重放 317 步，护架验收在回合末完成；采用视频最后可见帧 3.16 s 为近似到达时刻，不把刚开始收拳计为已完成连续保持。单视角不能精确测出掌肩距离、骨盆速度和连续 200 步计数，这些阈值由该交付文件的裁判成功记录确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "机器人收回右拳并保持护架时，左右脚持续与地面接触承重，双腿和躯干维持站立；视频 1.20–3.16 s 未见手或膝触地。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": 14.88
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": 0.056
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": 1.563
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人完成有效击袋和护架站稳两步：约视频 1.04 s 完成击打结算，右拳随后收回并站立至视频 3.16 s 的回合末验收。交付文件的官方重放与最终评分器均确认成功，有效冲量 14.88 N·s、接触历时 0.056 s、首次接触高度 1.563 m。无状态日志，视频无法独立精测拳峰接触位置、冲量及护架数值阈值；这些依裁判记录确认，节点时间为与视频核对后的近似值。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "成功运行；bag_struck（视频约1.04 s）与 guard_held（视频3.16 s）与帧、log step 30 的官方重放（strike 仿真1.028 s，success 3.17 s，317步）及评分器字段一致，contact_actual 合理，无失败标签，推断/观察标注正确。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:44:46Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--mpenplb/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--mpenplb/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "c474b21f98cf1ce65aa61a26bd3d676d5d09d64c7806d3c6f1a4b63692f936e7",
+   "video_sha256": "1c23b1c7ba19324690d7ca3f63177cbcc268a797400d4c3d742e0be3aaf24d13",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--sswikn5.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--sswikn5.json
new file mode 100644
index 0000000000000000000000000000000000000000..1fc4f08c70f2dffd8190f463406c15686b7b035b
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--sswikn5.json
@@ -0,0 +1,220 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-privileged--sswikn5",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 287,
+   "sim_steps": 287,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 3.9,
+   "bad_trajectory": 0,
+   "metric_contact_impulse_n_s": 25.39,
+   "metric_contact_s": 0.042,
+   "metric_hit_height_m": 1.412,
+   "metric_impulse_n_s": 15.62,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 276,
+   "agent_resets": 10889,
+   "agent_replays": 38,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-privileged--sswikn5/replay.mp4",
+  "duration_s": 2.88,
+  "fps": 25.0,
+  "frames": 72,
+  "sha256": "6fc8420c2ac1d3d87a574f6a2da3a34cff99b51ab8353a4686d88e5207449886"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 2,
+  "final": 2,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [
+     [
+      0.68,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.6,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.52 s 机器人在沙袋前站立并预备右臂，视频 0.56–0.64 s 机器人转动躯干、右拳向前上方击向红色沙袋的白色环带；视频 0.68–0.80 s 右肘弯曲、拳袋分离，沙袋随后摆离。双脚提供地面支撑，未见机器人跌倒。"
+     },
+     {
+      "t": 0.68,
+      "log_step": 594,
+      "text_zh": "观察：log step 593 重新生成最终交付文件，log step 594 对该文件的两次独立新进程重放均报告 stages=[strike,guard]、success=true、相同终态哈希；strike 事件记录在仿真 0.663 s，有效冲量 15.62 N·s、接触时长 0.042 s、首次击中高度 1.412 m，与本次评分 reward 字段一致。"
+     },
+     {
+      "t": 0.68,
+      "log_step": 594,
+      "text_zh": "推断：结合视频 0.60–0.68 s 的白带击打及分离和最终交付的裁判 strike 事件，将视频 0.68 s 记为有效击打已结算的近似时刻；精确冲量和拳峰合法性由本次成功裁判记录确认，不能由摆幅独立测定。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "观察：视频 0.60–0.64 s 机器人右手末端拳垫击中沙袋白带，随后收回；候选接触名为“击打”，现有怎么碰条目没有短促冲击，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–0.80 s 机器人脚接地承重，腿和躯干在预备、转体出拳与收拳时维持站立。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.87,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.88,
+      "log_step": null,
+      "text_zh": "观察：视频 0.68–0.88 s 机器人弯曲右肘将拳从沙袋前收至右肩附近；视频 0.88–2.84 s 保持该护架姿态、身体直立，仅脚接地且无明显整体平移，沙袋继续摆动；已逐帧查看视频末段 1.88–2.84 s，未见再触袋或倒地。"
+     },
+     {
+      "t": 2.84,
+      "log_step": 594,
+      "text_zh": "观察：log step 594 的最终交付重放报告在控制步 287 完成 guard，success=true、fell 对应本次评分字段为 0；本次 reward 同时报告 stages_reached=2、stages_total=2、replay_success=1、deterministic=1。"
+     },
+     {
+      "t": 2.87,
+      "log_step": 594,
+      "text_zh": "推断：log step 594 是最终交付文件的控制步计数；log step 7 所读控制周期为 0.01 s，287×0.01=2.87 s，与录像时长 2.88 s 相符。其 strike 事件仿真 0.663 s 已与视频 0.60–0.68 s 的击打、分离核对，故以终末控制步对应的视频 2.87 s 标记护架连续保持验收完成；最后可见帧为视频 2.84 s，终末验收比该帧晚约 0.03 s。精确肩掌距离、骨盆速度和连续 200 步判据依据成功裁判记录。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.88–2.84 s 机器人脚与地面承重接触，腿和躯干支撑收拳后的直立护架；右拳保持离开沙袋。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": 15.62
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": 0.042
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": 1.412
+  }
+ ],
+ "failure": null,
+ "summary_zh": "观察：本次评分录像中，机器人约在视频 0.60–0.64 s 用右拳击中沙袋白带，随后收拳至右肩附近并站立至视频末帧 2.84 s；最终交付的两次新进程重放及评分记录均确认击打、护架两节点完成，冲量 15.62 N·s、接触时长 0.042 s。推断：按最终交付控制步映射，视频约 0.68 s 完成击打结算，视频 2.87 s 完成护架连续保持验收，无失败断点。观察：没有状态日志，录像不能独立量出拳峰接触区、冲量及护架的精确距离和速度阈值，这些判据依据与录像一致的最终交付裁判记录。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "两个节点均达成，击打约视频 0.60–0.64 s（帧 15–16）、沙袋随后摆离，护架保持至末帧；控制步 287×0.01=2.87 s 的映射有说明且与视频时长相符；contact_actual 合理，无失败断点，未发现需改动之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:53:39Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--sswikn5/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--sswikn5/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "520aeba035ed3162b9d1377c35a0046558d57d4c203412956325b8e431850ee4",
+   "video_sha256": "6fc8420c2ac1d3d87a574f6a2da3a34cff99b51ab8353a4686d88e5207449886",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--sz2kvle.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--sz2kvle.json
new file mode 100644
index 0000000000000000000000000000000000000000..687659a4ea80fb39a460848d079a7fd4fbe8dbb9
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--sz2kvle.json
@@ -0,0 +1,222 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-privileged--sz2kvle",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 319,
+   "sim_steps": 319,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 4.43,
+   "bad_trajectory": 0,
+   "metric_contact_impulse_n_s": 18.3,
+   "metric_contact_s": 0.041,
+   "metric_hit_height_m": 1.392,
+   "metric_impulse_n_s": 15.23,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 42,
+   "agent_resets": 202,
+   "agent_replays": 12,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.691,
+   "peak_processes": 11,
+   "peak_disk_gb": 0.24,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-privileged--sz2kvle/replay.mp4",
+  "duration_s": 3.2,
+  "fps": 25.0,
+  "frames": 80,
+  "sha256": "d32a2dce5d6465a205977a4e28508cd938aa894599a56894192066c65188b225"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 2,
+  "final": 2,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [
+     [
+      1.16,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 1.08,
+      "log_step": null,
+      "text_zh": "观察：视频 1.00–1.12 s，机器人右臂向前快速伸展，右手末端到达红色沙袋白色环带；视频 1.16–1.28 s 拳袋分开，沙袋向远离机器人的方向摆动，右肘弯曲收拳。双脚持续接地支撑。"
+     },
+     {
+      "t": 1.16,
+      "log_step": 143,
+      "text_zh": "观察：inputs/run.json 的评分器字段 replay_success=1、deterministic=1、stages_reached=2/2，metric_impulse_n_s=15.23、metric_contact_s=0.041、metric_hit_height_m=1.392；log step 143 对最终交付文件的两次独立重放均由仿真裁判返回 strike、guard 阶段成功且终态哈希相同。"
+     },
+     {
+      "t": 1.16,
+      "log_step": null,
+      "text_zh": "推断：结合视频 1.08 s 的击袋及视频 1.16 s 后的拳袋分离与评分器记录，有效击打在视频约 1.16 s 完成；冲量和短接触是否达标以本次评分器字段为据，不能由摆幅独立测出。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "视频 1.08 s 附近，机器人右手末端拳垫击中沙袋白色环带，随后收回并与沙袋分离。候选名称“击打”：短促冲击不属于抓握、推/扫或压/擦。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 0.92–1.36 s 出拳与收拳过程中，机器人双脚接触地面承重，腿和躯干维持直立。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.19,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 1.28,
+      "log_step": null,
+      "text_zh": "观察：视频 1.20–1.36 s，机器人右肘弯曲，将出拳的右手收至右肩附近；视频 1.40–3.16 s 右臂保持回收，身体直立、双脚留在地面，沙袋往返摆动而未见再次触袋。视频 2.20–3.16 s 已逐帧检查，包括末帧。"
+     },
+     {
+      "t": 3.16,
+      "log_step": 143,
+      "text_zh": "观察：视频 3.16 s 末帧机器人仍维持护架站姿；本次评分器 success=1、stages_reached=2、fell=0。log step 143 的最终文件重放返回 stages=[strike, guard]、steps=319、time_s=3.19、failed=null，说明裁判已接受连续护架保持。"
+     },
+     {
+      "t": 3.19,
+      "log_step": 143,
+      "text_zh": "推断：log step 143 所报最终轨迹 319 个控制步对应仿真 3.19 s，即每步 0.01 s，与视频总长 3.20 s 相符；log step 139 对该文件诊断的接触控制步 107–110 对应视频约 1.07–1.10 s，已用视频 1.08 s 的白带接触核对。故将护架验收完成锚定为视频约 3.19 s，精度约一个视频帧；视频 1.20 s 附近开始收拳不等于已经完成连续保持验收。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 1.20–3.16 s，机器人以双脚接地支撑护架站立，腿和躯干维持直立，右手收于肩部附近；未见手、膝或躯干落地。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": 15.23
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": 0.041
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": 1.392
+  }
+ ],
+ "failure": null,
+ "summary_zh": "本次评分轨迹完成击中沙袋与护架站稳两节点：视频约 1.16 s 完成击打分离，随后收拳，视频约 3.19 s 连续护架验收完成；评分器记录冲量 15.23 N·s、接触历时 0.041 s、击中高度 1.392 m，成功且未跌倒。成功参考视频中击打后沙袋摆动与收拳站立的线索同样可见，本次节点仅依据本次视频及本次裁判结果判定。没有状态日志，精确拳峰接触位置、护架距离与速度阈值无法从画面独立量测，阈值是否满足依据评分器及最终文件重放成功记录，完成时刻有约一帧误差。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "两节点均达成、成功运行无失败层；视频约1.08–1.16 s击袋、收拳后站稳至3.19 s，与评分器字段一致，控制步到视频时间的换算有证据。未发现需改动之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:58:30Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--sz2kvle/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--sz2kvle/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "109b2e326044c9e16f196656b98ec7e07bb1e4dd67c7a1c7e9554a7e0d5cfdf0",
+   "video_sha256": "d32a2dce5d6465a205977a4e28508cd938aa894599a56894192066c65188b225",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--zxqrdj3.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--zxqrdj3.json
new file mode 100644
index 0000000000000000000000000000000000000000..afad08f7f4a846bdceca39ecc27a87838cedb701
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-privileged--zxqrdj3.json
@@ -0,0 +1,222 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-privileged--zxqrdj3",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 364,
+   "sim_steps": 364,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 3.3,
+   "bad_trajectory": 0,
+   "metric_contact_impulse_n_s": 15.41,
+   "metric_contact_s": 0.071,
+   "metric_hit_height_m": 1.456,
+   "metric_impulse_n_s": 13.87,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 80,
+   "agent_resets": 0,
+   "agent_replays": 4,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.6,
+   "peak_processes": 7,
+   "peak_disk_gb": 0.25,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-privileged--zxqrdj3/replay.mp4",
+  "duration_s": 3.68,
+  "fps": 25.0,
+  "frames": 92,
+  "sha256": "a806de4ff8c35f613b12bd58eef383edfad67be4ac316e6d0a2f8379fecb01df"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 2,
+  "final": 2,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [
+     [
+      0.56,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.44,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 0.36–0.68 s，机器人转动躯干、右臂前伸，视频 0.44–0.48 s 右手末端在沙袋白色环带处击打；随后右肘弯曲收拳，视频 0.56 s 拳袋已分离，沙袋开始摆离。双脚支承身体，未见其他部位落地。"
+     },
+     {
+      "t": 0.56,
+      "log_step": 94,
+      "text_zh": "观察：log step 94 对最终交付文件的重放报告 stages=[strike, guard]、success=true，strike 事件在仿真 0.537 s；评分器 reward 同时报 metric_impulse_n_s=13.87、metric_contact_s=0.071、metric_hit_height_m=1.456，分别满足冲量至少 12.0 N·s、接触不超过 0.10 s 和白带高度要求。"
+     },
+     {
+      "t": 0.56,
+      "log_step": 94,
+      "text_zh": "推断：最终文件重放的 364 个控制步对应仿真 3.64 s，即每步 0.01 s，与视频末帧 3.64 s 及视频 0.44–0.56 s 的击袋分离动作相符；据此将仿真 0.537 s 的 strike 结算定位于视频约 0.54 s，取已检查的邻近帧视频 0.56 s 为节点锚点。精确冲量与拳峰接触合法性由评分器通过结果支持，非从摆幅测得。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "观察：视频 0.44–0.48 s 机器人右手末端固定拳垫短促击中沙袋白带后收回；候选接触名为“击打”，现有怎么碰条目不能准确表示短促冲击，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.36–0.68 s 双脚与地面支承机器人出拳和收拳，腿与躯干维持身体直立。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.64,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.68,
+      "log_step": null,
+      "text_zh": "观察：视频 0.52–0.80 s 机器人右肘弯曲、右手从袋前收至肩旁；视频 0.80–1.64 s 机器人仍有小幅足部与身体调整，未将刚收拳的一帧计为连续站稳完成。"
+     },
+     {
+      "t": 3.64,
+      "log_step": 94,
+      "text_zh": "观察：逐帧检查视频 1.64–3.64 s，机器人保持右肘弯曲、右手在右肩附近、躯干直立，足部支撑且身体位置基本稳定；视频最后一秒 2.68–3.64 s 未见跌倒或再次触袋。log step 94 最终交付文件重放的 guard 与 success 事件均在仿真 3.64 s，reward 的 stages_reached=2、replay_success=1、deterministic=1、fell=0。"
+     },
+     {
+      "t": 3.64,
+      "log_step": 94,
+      "text_zh": "推断：最终文件的 364 控制步×0.01 s 与视频末帧 3.64 s 一致，并已用视频 0.44–0.56 s 的击袋动作核对时序；护架连续保持完成以视频 3.64 s 为锚点。成功裁判结果支持掌肩距离、骨盆速度及连续 200 控制步等无法仅凭画面精确量出的验收条件。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.68–3.64 s 机器人通过双脚接地及腿部小幅调整支撑收拳后的身体，后段保持直立护架；右手收在肩旁，与沙袋分离。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": 13.87
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": 0.071
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": 1.456
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人在视频约 0.44–0.48 s 用右手末端击中沙袋白带，分离后完成有效击打，随后收拳调整站姿，在视频 3.64 s 完成连续护架站稳，最终两节点均达成。最终交付文件重放及评分器均判成功，冲量 13.87 N·s、接触时长 0.071 s、击中高度 1.456 m 与视频动作一致。已检查本次全部 92 帧及参考成功视频；无状态日志，精确冲量、拳峰接触位置和护架阈值依评分器及最终重放记录确认，节点时间误差约一帧至 0.1 s。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "成功运行，两节点最终达成；视频 0.44 s 右拳触袋、3.64 s 护架站稳与评分器结果一致，接触标签合理，无失败层。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:51:16Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--zxqrdj3/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-privileged--zxqrdj3/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "4fe3dc1993bcaa97c5d2a2bef70cde1c0c75a46b07e51bde4ab5e4bf4fa83162",
+   "video_sha256": "a806de4ff8c35f613b12bd58eef383edfad67be4ac316e6d0a2f8379fecb01df",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--hvtgpyv.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--hvtgpyv.json
new file mode 100644
index 0000000000000000000000000000000000000000..4432b52c0a4463a9c98f3b1be9cf9e90812ecf07
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--hvtgpyv.json
@@ -0,0 +1,221 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-standard--hvtgpyv",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 42,
+   "sim_steps": 257,
+   "calls": 15,
+   "rows": 257,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 487.9,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.213,
+   "peak_processes": 7,
+   "peak_disk_gb": 0.281,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-standard--hvtgpyv/replay.mp4",
+  "duration_s": 0.44,
+  "fps": 25.0,
+  "frames": 11,
+  "sha256": "9a636ff4f0fb4e49632b7dcc9d2c33caa89d4e0f4be89d61447c5f709400afec"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": 111,
+      "text_zh": "观察：已逐帧检查视频 0.00–0.40 s 的全部 11 帧；机器人双脚站立，先弯肘蓄拳，视频 0.36–0.40 s 右手向红色沙袋白带伸出，末帧右手仍贴近袋面。log step 111 显示最后一次 reset 后执行 episode 6，机器人服务返回 done=True、reason=failed、stages_reached=0/2，并由内置裁判明确判定“the bag was touched by something other than the knuckle part of the right fist”；此为服务裁判结果，非 agent 自述。有效拳峰击打未获验收。"
+     },
+     {
+      "t": 0.4,
+      "log_step": 112,
+      "text_zh": "观察：log step 112 确认最终 episode=6、control_steps=42、done=True；此前 episode 1–5 均被 reset，不将其击打或失败计入本视频。视频仅有最后一次蓄拳、伸拳，未显示成功参考视频中击后分离、收拳并持续护架的后续过程。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "观察：视频 0.36–0.40 s 机器人右手末端向沙袋白带出拳，log step 111 的裁判结果确认发生非合法拳峰部位触袋；候选接触名“击打”，短促出拳接触不属于抓握、推/扫或压/擦。画面不能分辨具体违规接触斑块。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–0.40 s 机器人双脚持续接地支撑蓄拳和出拳，未见跌倒。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": 111,
+      "text_zh": "观察：视频 0.00–0.40 s 全程未出现有效击打后收回右拳并护架保持的过程，末帧右臂仍伸向袋面；log step 111 的最终回合由裁判因违规触袋终止，stages_reached=0/2。推断：击打前置未完成，且整个视频不足要求的击后连续 2.0 s，护架站稳节点从未成立；脚本中尚未执行完成的护架计划不计为进度。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bag_struck",
+   "t": 0.4,
+   "log_step": 111,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 0.36–0.40 s 机器人正在把右手伸向白带，尚未完成拳袋分离和护架；log step 111 的最终 episode 6 被裁判以非右拳拳峰部位触袋判失败，stages_reached=0/2。断点属于正在进行的击中沙袋步骤；视频 0.40 s 为可见末段接触的近似导航锚点，未把 agent 工作时间换算成视频时间。"
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 0.4,
+    "log_step": 111,
+    "evidence_zh": "观察：视频 0.36–0.40 s 机器人右手伸至沙袋白带附近；log step 111 的内置裁判报告实际触袋者包含合法右拳拳峰之外的部位。推断：这次出拳没有把实际接触限制在要求的拳峰区域，属于击打接触位置或姿态不准；并非完全没有触袋，也未见机器人失衡。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 0.4,
+    "log_step": 110,
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "观察：log step 105 的离线运动学计算报告目标拳峰位置及掌部间隙，log step 110 的最终脚本按固定关节插值执行蓄拳、快速伸拳与回收，并声称目标姿态可保证拳峰接触；视频 0.36–0.40 s 显示机器人确实伸向白带，但 log step 111 的实际执行被裁判判为非拳峰触袋。推断：接触控制未将目标点上的静态校验落实为合法的实际接触；证据不足以继续区分接触几何设计或动态跟踪问题，故停留在“控制”。"
+   }
+  ],
+  "gap_zh": "没有状态日志及逐接触部位记录；视频 0.36–0.40 s 的远景不足以确定具体是哪一块非拳峰表面触袋，违规事实依据最终 episode 6 的服务裁判反馈。评分器未报告 impulse_n_s、contact_s、hit_height_m，三项原生指标均留空；不能从 agent 在 log step 114 所称“沙袋变形”确定材料变形机制或控制子类。",
+  "unmet": {
+   "bag_struck": false,
+   "guard_held": false
+  }
+ },
+ "summary_zh": "评分采用最后一次现场 episode 6：机器人站立蓄拳并向沙袋白带出拳，但裁判因非右拳拳峰部位触袋终止回合，击中沙袋与护架站稳均未完成（0/2）。断点在击中沙袋操作中，表现为实际击打接触位置或姿态不准，根因保守标为控制。已检查视频全部 11 帧并对照成功参考；无状态日志，具体违规接触斑块和动态误差来源无法细分，裁判也未报告三项原生指标。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "两节点未达成、断点在击中沙袋（操作中）、位姿不准+控制（保守）均有证据支持：视频 0.00–0.40 s（11 帧）仅见蓄拳与伸向沙袋，末帧手掌贴袋；log step 111 服务裁判返回 reason=failed、stages_reached=0/2、非右拳拳峰部位触袋。时钟均为视频时间，未换算 agent 时间；未发现需修改项。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:52:00Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--hvtgpyv/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--hvtgpyv/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "666579a6a337f9a36a63fcf30be515ae6dcd77bcf73c55fb810eaa4ed4bcdcfb",
+   "video_sha256": "9a636ff4f0fb4e49632b7dcc9d2c33caa89d4e0f4be89d61447c5f709400afec",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--nbxbqbe.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--nbxbqbe.json
new file mode 100644
index 0000000000000000000000000000000000000000..0b69f0ad3f9800be7fdfb6b6eb72524a3afd64b0
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--nbxbqbe.json
@@ -0,0 +1,214 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-standard--nbxbqbe",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 343,
+   "sim_steps": 1402,
+   "calls": 36,
+   "rows": 19,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2073.3,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 3,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.21,
+   "peak_processes": 12,
+   "peak_disk_gb": 1.93,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-standard--nbxbqbe/replay.mp4",
+  "duration_s": 3.44,
+  "fps": 25.0,
+  "frames": 86,
+  "sha256": "648cc20be7fbf75154bfb47121c814d5ec3ca803eeaffaef9d9a4225cd356117"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 136,
+      "text_zh": "观察：本次视频从机器人站在沙袋前开始，右臂先抬起、屈肘并分段调整；视频 0.00–3.40 s 的沙袋基本竖直，未见成功参考中击打后明显摆离、拳袋分离的过程。视频 2.32–3.40 s 已逐帧检查，机器人没有跌倒。"
+     },
+     {
+      "t": 3.4,
+      "log_step": 136,
+      "text_zh": "观察：log step 136 明确运行最后一次现场回合，重置后 resets_left=0；guard、wind1、wind2、windF 均返回未结束，punch 返回 done=True、reason=failed。log step 137 的服务状态为 episode=6、control_steps=343，未打印 strike 事件或冲量。"
+     },
+     {
+      "t": 3.4,
+      "log_step": 137,
+      "text_zh": "推断：log step 11 给出控制周期 0.01 s；log step 130 的最终流程在出拳前依次执行 1.2+0.9+0.6+0.7=3.40 s，且 log step 136 未执行额外 windG。该控制步时钟与视频约 1.20 s 的右臂姿态切换及视频末端吻合；343 步对应约视频 3.43 s，出拳仅起动约 0.03 s 即判负。结合未见沙袋摆离与拳袋分离，未完成符合冲量、接触及部位要求的有效击打，不能凭 agent 的本地模拟结果记为达成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–3.40 s，机器人双脚接地，腿和躯干支撑右臂准备及出拳起动，全程未见倒地。"
+     },
+     {
+      "who": "未知",
+      "how": "未知",
+      "note_zh": "观察：视频末端可见机器人右臂朝沙袋白带方向起动出拳，但固定拳垫的真实接触面细小，视频在起动处结束，无法确认是否发生触袋及首先触袋的具体部位；此处标的是目标接触未知，出拳步骤确已尝试。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.4,
+      "log_step": 136,
+      "text_zh": "观察：视频 0.00–3.40 s 只显示击打前的抬臂、屈肘准备和末端出拳起动，未显示有效击打之后的收拳及连续站稳保持。log step 136 在 punch 即返回失败，没有执行后续 retract、guard_hold；开局的护架准备不能作为击打后的完成状态。"
+     },
+     {
+      "t": 3.4,
+      "log_step": 137,
+      "text_zh": "推断：前置 bag_struck 未达成，且视频结束前不存在击打后的连续 2.0 s 护架窗口，因此 guard_held 从未达成；本回合未开始该后续步骤。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bag_struck",
+   "t": 3.4,
+   "log_step": 136,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 3.40 s，机器人仍站立，右臂从袋前屈肘姿态起动出拳，沙袋未出现明显摆离；log step 136 的最后现场回合在 punch 返回 failed，log step 137 确认回合已结束。推断：按控制周期及该回合命令长度，终止发生在出拳起动约 0.03 s 后；未修复的失败步骤为有效击中沙袋，随后护架尚未开始。"
+  },
+  "unmet": {
+   "bag_struck": false,
+   "guard_held": false
+  },
+  "phenomena": [],
+  "root_causes": [],
+  "gap_zh": "观察：inputs/ 有完整视频和会话日志，但没有状态日志；最后回合的日志只打印 reason=failed，未打印裁判 outcome、stages_reached、strike、impulse_n_s、contact_s 或接触点。评分字段 first_reason_code=3 未附代码含义，不能自行解码。视频最后一秒已逐帧检查，但 25 fps 的场馆远景不能分辨出拳起动末端约 0.03 s 内的拳峰、掌侧或前臂触袋先后。推断：因此可确定断点与两节点未达成，不能可靠区分部位或位姿违规、力不足等具体失败现象，也不能在接触设计、轨迹跟踪或感知误差之间确定根因，现象与根因均留空。agent 在 log step 138、140 将最后回合描述为“可能推袋超时或接触姿态违规”，属于其推测；此前重置回合的推袋与本地模拟冲量不是本次回合的事实。三个原生指标均未由评分器报告，故 value=null。"
+ },
+ "summary_zh": "观察：最后现场回合对应本次视频，机器人完成右臂抬起及分段准备，在视频约 3.40 s 起动出拳后被服务判负，沙袋未见明显摆离，有效击袋与击打后的护架保持均未达成，最终为 0/2。推断：断点为 bag_struck 的操作中；裁判细项与接触状态未记录，末端短暂接触亦无法从远景分辨，因此未采纳 agent 关于推袋超时的猜测，现象及根因留空。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "两节点未达成、断点 bag_struck 操作中、控制步时钟（343 步≈3.43 s，与视频 3.44 s 吻合）均有帧和 log step 136/137 支持；现象与根因留空并写明 gap_zh，因裁判细项缺失而无法区分，符合证据纪律。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:51:32Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--nbxbqbe/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--nbxbqbe/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "02b490c94d0c0b583a2571a969999f09e0f58afc511d095b5a7f67ea71ad28e3",
+   "video_sha256": "648cc20be7fbf75154bfb47121c814d5ec3ca803eeaffaef9d9a4225cd356117",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--odhv9c9.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--odhv9c9.json
new file mode 100644
index 0000000000000000000000000000000000000000..e72105d9a0320091d6ddcaac39f5afd6eed1a171
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--odhv9c9.json
@@ -0,0 +1,216 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-standard--odhv9c9",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "live_success": 1,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 417,
+   "sim_steps": 771,
+   "calls": 21,
+   "rows": 771,
+   "resets": 2,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 370.6,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.33,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.842,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-standard--odhv9c9/replay.mp4",
+  "duration_s": 4.2,
+  "fps": 25.0,
+  "frames": 105,
+  "sha256": "f8356ef83e80eb0e2afa5caecb1659d425d3ffd7112c33095b21bd1304d717a2"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 2,
+  "final": 2,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [
+     [
+      1.2,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 1.04,
+      "log_step": 38,
+      "text_zh": "观察：视频 0.92–1.12 s，机器人右臂迅速向前伸出，右手末端在沙袋白色环带处接触；视频 1.16–1.28 s 右肘弯曲、拳袋分离，沙袋随后摆离。视频全程只见这一次击袋动作。"
+     },
+     {
+      "t": 1.2,
+      "log_step": 45,
+      "text_zh": "推断：以视频约 1.20 s 作为拳袋分离后的击打结算锚点，误差约 0.1 s；log step 45 中机器人服务的末回合裁判返回 episode=3、outcome=success、stages_reached=2、stages_total=2，结合视频可判此击被裁判接受，包括带内拳峰、冲量和接触时长判据；不能由摆幅独立测出精确冲量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "观察：视频约 1.04–1.12 s，机器人右手末端拳垫击打红色沙袋的白色环带后迅速分离。候选名称“击打”；现有方式条目没有短促冲击，不属于抓握或持续推扫。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0–1.20 s，机器人双脚落地支撑出拳，腿和躯干保持站立，未见膝或躯干落地。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.17,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 1.28,
+      "log_step": 38,
+      "text_zh": "观察：视频 1.16–1.28 s，机器人右拳从袋前收回肩前附近；视频 1.28–2.16 s 躯干有小幅前后调整，仍保持站立。视频 2.20–4.16 s 右肘持续弯曲、右拳保持肩前，身体趋于直立、位置变化很小，未见再次触袋或跌倒；末秒视频 3.20–4.16 s 已逐帧核对。"
+     },
+     {
+      "t": 4.17,
+      "log_step": 45,
+      "text_zh": "观察：log step 45 的机器人服务返回末回合 episode=3、control_steps=417、sim_time_s=4.17、done=true、reason=success、stages_reached=2/2；这是裁判现场结果。log step 7 的服务规格给出 control_dt=0.01 s。"
+     },
+     {
+      "t": 4.17,
+      "log_step": 45,
+      "text_zh": "推断：采用本次评分回合的控制步映射，417×0.01=视频约 4.17 s，作为连续护架站稳验收完成时刻；log step 38 的第 3 回合出拳结束 control_steps=106 对应视频约 1.06 s 的伸拳、随后收拳，且末回合 4.17 s 与视频末帧 4.16 s 对齐，已核对映射。精确连续保持由成功裁判确认，不能仅以视频中开始收拳的时刻提前记为完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 1.28–4.16 s，机器人双脚支撑地面，腿和躯干调整姿态并保持直立；右拳收于肩前，不再接触沙袋。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": null,
+ "summary_zh": "观察：评分视频对应 log step 37 重置后的第 3 回合；机器人约在视频 1.04–1.12 s 以右拳击中白带，约在视频 1.20 s 完成击打分离，随后收拳并站立保持，log step 45 的现场裁判确认 2/2 成功。推断：按末回合控制步数与视频核对，将护架站稳验收记在视频约 4.17 s；前两回合的带外击打和跌倒仅属会话调试。观察：本次没有状态日志，评分记录和服务输出未报告 impulse_n_s、contact_s、hit_height_m 数值，故原生指标均留 null，精确冲量、拳峰接触位置及连续保持判据依赖末回合裁判成功结果。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "评分回合成功（2/2），两节点时间和接触标注与视频及 log step 38/45 的控制步映射一致（417×0.01=4.17 s，视频 4.2 s），未发现需修改之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:59:50Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--odhv9c9/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--odhv9c9/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "ec194e76eef3e88e8a8ff3b88400f1c7c99a86e66a1c2354cbdbbcbfd1253eae",
+   "video_sha256": "f8356ef83e80eb0e2afa5caecb1659d425d3ffd7112c33095b21bd1304d717a2",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--pucvb24.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--pucvb24.json
new file mode 100644
index 0000000000000000000000000000000000000000..1dd52a400f9fe06855baa43564514f1fd70485c5
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--pucvb24.json
@@ -0,0 +1,258 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-standard--pucvb24",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 208,
+   "sim_steps": 287,
+   "calls": 6,
+   "rows": 287,
+   "resets": 2,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 7055.3,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.223,
+   "peak_processes": 8,
+   "peak_disk_gb": 0.311,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-standard--pucvb24/replay.mp4",
+  "duration_s": 2.12,
+  "fps": 25.0,
+  "frames": 53,
+  "sha256": "a3ce80b1b6493f6d86672a889bc4804af998a0e6cf75922672ecb49dccec03b4"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [
+     [
+      0.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.56,
+      "log_step": null,
+      "text_zh": "观察：视频 0.36–0.52 s 机器人右手末端在画面中已贴靠沙袋白色环带，视频 0.56–0.64 s 沙袋摆离、右手撤开，视频 0.72 s 拳袋已明确分离；双脚接地支撑。视频不能独立测量实际接触历时与冲量，是否达到击打判据以 log step 251 的末回合裁判 stages_reached=1 结合视频判定。"
+     },
+     {
+      "t": 0.72,
+      "log_step": 251,
+      "text_zh": "推断：log step 251 的现场服务记录为最后一次重置后的 episode 3，裁判返回 stages_reached=1、stages_total=2、outcome=fell，与视频中的击袋后后倒一致，故有效击打已经通过裁判验收；首次完成约取视频 0.72 s，视觉定位误差约 0.1 s。该记录的 208 个控制步、dt=0.01 s 对应仿真 2.08 s，已与视频 2.08 s 的倒地末帧核对；不使用 agent 工作时间换算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "观察：视频 0.52–0.60 s 机器人右手末端拳垫击中沙袋白带，随后撤离。候选接触名“击打”，现有接触方式无短促冲击条目，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.48–0.72 s 机器人左右脚接地，双腿支承身体完成出拳和撤拳。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.92,
+      "log_step": 250,
+      "text_zh": "观察：视频 0.76–0.92 s 机器人右肘弯曲，右手从袋前收至右肩附近，开始护架；log step 250 的脚本也在击打及撤拳之后切换到护架关节目标。"
+     },
+     {
+      "t": 1.6,
+      "log_step": 251,
+      "text_zh": "观察：视频 1.20–1.56 s 机器人身体逐渐向后倾斜，视频 1.60–2.08 s 后仰明显加剧直至倒下，未在击打后连续站稳 2.0 s。log step 251 的现场裁判仅计 1/2 阶段，终止原因为 fell。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.92–1.56 s 机器人双脚接地，腿和躯干尝试维持收拳护架；视频 1.60–2.08 s 脚部支承未能维持平衡，身体向后倒下。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "guard_held",
+   "t": 1.6,
+   "log_step": 251,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 0.92 s 机器人已收拳护架，此后持续后仰，视频 1.60 s 起明显失稳且未恢复，至视频 2.08 s 倒下；log step 251 的最后现场回合由裁判以 fell 终止，护架阶段未通过。"
+  },
+  "unmet": {
+   "bag_struck": true,
+   "guard_held": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 1.6,
+    "log_step": 251,
+    "evidence_zh": "观察：视频 1.60–2.08 s 机器人上体与骨盆向后下方运动，身体由站立转为向后倒下；log step 251 返回裁判 outcome=fell。失稳发生在击袋完成后的护架保持期间。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：log step 250 的实际执行脚本一次性提交按时间插值的 stand 关节目标，击打后转至固定护架姿态，没有根据回传姿态调整后续指令；视频 0.92–2.08 s 机器人在此阶段持续后倾且未纠正。log step 251 确认该脚本的最后回合以跌倒结束，支持击打后站立控制未能稳定身体；尚不能细分为具体增益、支承布局或反馈环节的问题。",
+    "source_zh": "日志"
+   }
+  ],
+  "gap_zh": "本次输入没有状态日志；评分器 reward 与最后回合服务结果未报告 impulse_n_s、contact_s、hit_height_m 数值，三个原生指标留空，有效击打以末回合裁判 stages_reached=1 结合视频判定。录像可见后倒，但不足以分辨具体失稳力矩、关节增益或支承控制缺陷，根因仅标控制。log step 86、134 属于此前两个已重置回合，其弱击打和违规触袋不是本段视频中的失败。"
+ },
+ "summary_zh": "最后评分回合约在视频 0.72 s 完成有效击袋，右拳随后收至肩旁，最终完成 1/2 节点。机器人在护架期间逐渐后仰，视频 1.60–2.08 s 失稳倒下，断于护架站稳；执行脚本的击打后站立控制未能维持平衡。无状态日志及精确冲量等原生读数，击打合格由末回合裁判阶段数与视频共同确认，具体失稳机制仍不明。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 7,
+    "summary_zh": "节点、时间、断点（guard_held，倒地）、平衡现象和控制根因均与视频及末回合裁判读数（stages_reached=1/2, outcome=fell）一致；仅 bag_struck 的第一条观察描述与帧不符。",
+    "n_issues": 1,
+    "n_major": 0
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "bag_struck 约视频 0.72 s 完成（末回合裁判 stages_reached=1/2），guard_held 未完成；断点 guard_held、现象平衡、根因控制均与视频 1.2–2.08 s 后倒及 log step 251 outcome=fell 相符；时钟换算（208 控制步=2.08 s=视频末帧）已核对，未发现需改动之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:02:41Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--pucvb24/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--pucvb24/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--pucvb24/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--pucvb24/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "979f57abe23d6748fba9a34cdec3d6211d85019fb5d5ee2c18249510cd11c452",
+   "video_sha256": "a3ce80b1b6493f6d86672a889bc4804af998a0e6cf75922672ecb49dccec03b4",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--q2mzaun.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--q2mzaun.json
new file mode 100644
index 0000000000000000000000000000000000000000..1b0baa11abe0edc6fb841dae24aa148dfa9a2ea5
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--q2mzaun.json
@@ -0,0 +1,225 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-standard--q2mzaun",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 100,
+   "sim_steps": 3056,
+   "calls": 134,
+   "rows": 1642,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2252.7,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.265,
+   "peak_processes": 11,
+   "peak_disk_gb": 1.07,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-standard--q2mzaun/replay.mp4",
+  "duration_s": 1.04,
+  "fps": 25.0,
+  "frames": 26,
+  "sha256": "c7e93ecc649d688efea30bae2f901a134f304c9bf92ff323b81fb05297c7cf07"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": 170,
+      "text_zh": "观察：视频 0.00–0.88 s 机器人站立、右臂准备，视频 0.92–1.00 s 右手向沙袋挥出，末帧右手到达白带下缘附近；未见击打后分离、沙袋明显摆离的后续过程。log step 170 的最后现场回合 episode=6 在 control_steps=100 时由服务裁判返回“the bag was touched by something other than the knuckle part of the right fist”，stages_reached=0；这是服务裁判结果，不是 agent 自述。"
+     },
+     {
+      "t": 1.0,
+      "log_step": 10,
+      "text_zh": "推断：log step 10 的服务规则明确仅最后一个现场回合受评；log step 170 的最后回合出拳终止与视频末尾动作一致。非法部位先触袋使有效击打未被接受，故本节点全程未达成，而非由总分为零单独推得。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "观察：视频 0.92–1.00 s 右手末端挥向沙袋并接触，log step 170 的裁判确认存在非右拳拳峰的违规触袋；候选接触名“击打”，短促挥击不适用抓握或推/扫。画面不能细分究竟是拳垫哪一面或邻近部位先接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–1.00 s 机器人双脚持续接地，双腿与躯干支撑准备和出拳，未见跌倒。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": 170,
+      "text_zh": "观察：视频 0.92–1.00 s 机器人仍在向袋挥右臂，视频结束前没有击打后收拳至肩旁并连续保持的阶段。log step 170 的服务在出拳阶段即返回 failed、stages_reached=0；脚本仅在未终止时才执行后续收拳和护架，此分支没有执行。"
+     },
+     {
+      "t": 1.0,
+      "log_step": 170,
+      "text_zh": "推断：前置有效击打未达成，且本回合在收拳护架前终止，无法满足有效击打后的连续 2.0 s 护架验收。开局站立不算完成此 final 节点。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "value": null,
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better"
+  },
+  {
+   "name": "contact_s",
+   "value": null,
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better"
+  },
+  {
+   "name": "hit_height_m",
+   "value": null,
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bag_struck",
+   "t": 1.0,
+   "log_step": 170,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 0.92–1.00 s 机器人右手向袋前上方挥出并到达袋面，身体仍站立；log step 170 的最后现场回合裁判在出拳时以“非右拳拳峰部位触袋”判负，stages_reached=0。log step 172 确认该回合已结束且不能再次重置，故击打步骤没有修复。"
+  },
+  "unmet": {
+   "bag_struck": false,
+   "guard_held": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 1.0,
+    "log_step": 170,
+    "evidence_zh": "推断：视频 0.92–1.00 s 的右手挥击到达沙袋，但 log step 170 的现场服务裁判判定接触不属于要求的右拳拳峰。问题是触袋时的接触部位/姿态未对准指定拳峰区；画面支持挥击触袋，不能独立分辨违规接触面的细节。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "抓取和接触设计",
+    "alt_feasible": "未知",
+    "evidence_zh": "推断：log step 170 的脚本预先将躯干偏航设为 0.30 rad，随后主要通过右肩俯仰从 0.4 rad 向 −0.90 rad 挥臂，其余右臂目标保持零；该接触姿态在视频 1.00 s 到达袋面时，被现场服务裁判判为非右拳拳峰触袋。所选挥击的接触几何未使规定拳峰独占触袋；具体应如何调整腕/臂姿态及可行替代未在本回合验证。",
+    "source_zh": "日志"
+   }
+  ],
+  "gap_zh": "本回合无状态日志；已逐帧检查视频 0.00–1.00 s，并对视频 0.84–1.00 s 放大检查，仍不能精确识别裁判判负的接触面，也不能测得冲量、接触时长或首次接触点高度。run.json 未报告 impulse_n_s、contact_s、hit_height_m，故原生指标留空。成功参考视频仅用于辨认击打后袋摆动和收拳护架的外观，不作为本回合证据；不采用 agent 关于“手指先碰”或“握拳无效”的猜测作为已证实的机械原因。"
+ },
+ "summary_zh": "评分对应 log step 170 的最后现场回合，视频中机器人站立准备后挥右手触袋，裁判以非右拳拳峰部位触袋终止；有效击打与护架站稳均未完成，最终进度 0/2。断点为击中沙袋，所选挥击接触姿态未满足拳峰限定；无状态日志，具体违规接触面及冲量等精确指标不可核实，先前重置回合的表现不计入本轨迹。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标签与视频（仅1.04 s，最后现场回合）及 log step 170/172 一致：服务裁判以非右拳拳峰触袋判负，stages_reached=0，两节点未达成，断点 bag_struck 合理；现象与根因有日志依据且标为推断，未发现需改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:56:51Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--q2mzaun/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--q2mzaun/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "9efdee9a73b99be97d55a49afade242d37213c0f2972f2222811114fdb9cf902",
+   "video_sha256": "c7e93ecc649d688efea30bae2f901a134f304c9bf92ff323b81fb05297c7cf07",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--uzi6vau.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--uzi6vau.json
new file mode 100644
index 0000000000000000000000000000000000000000..a7ee64c2f3cacfefeaaa8e1382f93b9c3229f05b
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--uzi6vau.json
@@ -0,0 +1,220 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-standard--uzi6vau",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 60,
+   "sim_steps": 378,
+   "calls": 12,
+   "rows": 378,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 5450.6,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.248,
+   "peak_processes": 12,
+   "peak_disk_gb": 2.975,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-standard--uzi6vau/replay.mp4",
+  "duration_s": 0.64,
+  "fps": 25.0,
+  "frames": 16,
+  "sha256": "309809651b52fddc3a00eaca44590e1eb9158c06d9ee3d8d0d37291fd185e566"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.52,
+      "log_step": null,
+      "text_zh": "观察：已逐帧检查视频 0.00–0.60 s 的全部 16 帧；机器人双脚支撑地面，先抬臂蓄力，再在视频 0.52–0.60 s 将右手迅速伸向沙袋白带，末帧右臂仍在袋前，未出现完成击打后分离的过程。"
+     },
+     {
+      "t": 0.6,
+      "log_step": 156,
+      "text_zh": "观察：log step 156 的机器人服务返回 reset 后 episode=6、control_steps=60、done=True、stages=0/2，并由现场裁判报告“the bag was touched by something other than the knuckle part of the right fist”；log step 158 确认最后仍是 episode=6、failed。视频 0.60 s 的右臂前伸触袋与此失败回合相符，因此这是违规击袋尝试，未完成有效击中沙袋节点；前五次重置测试不计入本视频进度。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "观察：视频 0.52–0.60 s 机器人右手末端朝沙袋白带快速击出；log step 156 的现场裁判确认触袋含右拳拳峰之外部位。候选接触名为“击打”，现有怎么碰条目没有短促冲击，不能标抓握；录像不足以辨明具体违规接触点。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–0.60 s 机器人双脚接地，腿和躯干支撑蓄力与出拳，未见跌倒。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.6,
+      "log_step": 156,
+      "text_zh": "观察：视频 0.52–0.60 s 机器人右臂继续向袋伸出，最后一帧未收回右肩附近，整段视频没有击打后连续 2.0 s 的护架保持。log step 156 的现场裁判在违规触袋时终止回合，stages=0/2；有效击打这一前置也未完成，护架步骤尚未开始。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bag_struck",
+   "t": 0.6,
+   "log_step": 156,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 0.52–0.60 s 机器人正向沙袋白带出拳，尚未收拳；log step 156 的最后一次现场回合 episode=6 在 control_steps=60 因右拳拳峰之外部位触袋而立即失败，stages=0/2。视频止于这次击打，未有后续修复。"
+  },
+  "unmet": {
+   "bag_struck": false,
+   "guard_held": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 0.6,
+    "log_step": 156,
+    "evidence_zh": "观察：视频 0.56–0.60 s 机器人右手及前臂伸至沙袋白带，末帧右臂仍伸出；log step 156 的现场裁判明确报告触袋部位不全是右拳拳峰。推断：击打时手臂与袋的相对位姿未把接触限制在规定拳峰区，属于击打接触位姿不准；不再将同一次违规接触重复标为违反约束。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "log_step": 156,
+    "evidence_zh": "推断：agent 的关键帧出拳控制未能在现场执行中维持规定的接触部位。log step 151、155 将出拳姿态、跟随段及驱动力矩写入脚本，log step 156 实际执行该脚本后，机器人服务报告非右拳拳峰触袋并终止，视频 0.52–0.60 s 同时显示右臂向袋击出。agent 在 log step 159 将原因解释为较强驱动和离线手部模型差异，但该细分解释缺少真实接触坐标验证，因此仅归为控制。",
+    "source_zh": "日志"
+   }
+  ],
+  "gap_zh": "无状态日志或逐接触记录；视频仅有场馆侧视且为 25 fps，无法分辨拳垫末端 0.03 m 的接触边界、具体违规部位及毫秒级接触起止。inputs/run.json 的奖励字段及最后一次服务输出未报告 impulse_n_s、contact_s、hit_height_m，故原生指标留 null；离线副本的数值不是本次现场回合的指标。尚不能区分接触几何设计、轨迹跟踪或力矩参数的具体贡献，agent 对模型差异和接触软硬的解释待核实。"
+ },
+ "summary_zh": "评分视频对应最后一次现场回合 episode=6：机器人站立蓄力后向沙袋白带出拳，但现场裁判因右拳拳峰之外部位触袋立即判负，有效击打与后续护架均未完成，最终为 0/2。断点在击打操作中，现象为接触位姿不准，根因归于出拳控制；录像无法定位具体违规接触点，原生冲量、接触时长和击中高度未报告。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标签与证据一致：视频仅 0.64 s（16 帧）对应 episode 6 的 60 个控制步，末帧右臂已触及沙袋；log 步骤 id 156 的服务读数为 failed、stages 0/2，两个节点均为 false，断点 bag_struck 操作中合理，现象和根因均已标为推断并说明了不确定性。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:00:17Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--uzi6vau/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--uzi6vau/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "7d8cf706b07a19778b4ccc2b34ca6b9f6a35a0fac825856a0cf677039b2b67f2",
+   "video_sha256": "309809651b52fddc3a00eaca44590e1eb9158c06d9ee3d8d0d37291fd185e566",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--vwthikh.json b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--vwthikh.json
new file mode 100644
index 0000000000000000000000000000000000000000..d4e5568a626961945c09e095608aa7c828c8dd98
--- /dev/null
+++ b/phase2/embodied-olympics-fun/boxing/trajectories/embodied-olympics-fun-boxing-i00-standard--vwthikh.json
@@ -0,0 +1,226 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-boxing-i00-standard--vwthikh",
+ "taskKey": "embodied-olympics-fun/boxing",
+ "dag_sha256": "5fe95f4fd88e6dbc49fa1e5ba7230e5fd61571eeb504fd1c0835f0ca7840e98d",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 226,
+   "sim_steps": 1677,
+   "calls": 296,
+   "rows": 1677,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 3903.2,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 3,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 4.445,
+   "peak_processes": 26,
+   "peak_disk_gb": 0.722,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-boxing-i00-standard--vwthikh/replay.mp4",
+  "duration_s": 2.28,
+  "fps": 25.0,
+  "frames": 57,
+  "sha256": "b23ee25775893e00fd1103b51c8c8470d4abfd35cad9215bbd3fa4ff8dd7060f"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bag_struck": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 150,
+      "text_zh": "观察：逐帧检查视频 0.00–2.24 s，机器人先抬起并调整右手，视频 2.00–2.24 s 右手向沙袋白带加速前伸，末帧手部到达袋面附近；没有随后拳袋分离、沙袋明显摆离的完整有效击打过程。log step 150 的最后一局 episode 6 由现场裁判返回 outcome=\"the bag was touched by something other than the knuckle part of the right fist\"、stages_reached=0、stages_total=2，故该次接触未被接纳为有效击打。"
+     },
+     {
+      "t": 0,
+      "log_step": 37,
+      "text_zh": "观察：log step 37 的服务规则明确只有最后一局受评；log step 150 重置进入 episode 6，log step 151 确认该局在 control_steps=226 时 failed。视频仅显示一次准备和末尾出拳，与该局相符；更早几局及本地仿真测试不计为本视频的进度。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "候选名称：击打。视频 2.16–2.24 s 右手末端向沙袋白带出击，log step 150 的现场裁判确认发生包含非右拳拳峰部位的触袋；不能从该视角区分具体是手指还是拳垫其他区域。属于单手短促击袋尝试，现有接触方式无击打项，不是抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–2.24 s，机器人双脚接地支承抬臂和出拳，腿、躯干保持站立，未见跌倒。"
+     }
+    ]
+   },
+   "guard_held": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 150,
+      "text_zh": "观察：视频 2.00–2.24 s 机器人右手仍向袋前伸，视频在这次接触附近结束，没有击打后收回右拳并连续护架站稳的过程。log step 150 中脚本发送恢复段后立即收到 failed、stages_reached=0/2，未进入有效击打后的保持阶段。"
+     },
+     {
+      "t": 1.6,
+      "log_step": null,
+      "text_zh": "推断：视频 0.80–2.00 s 的屈肘站立是击打前准备，bag_struck 尚未达成，不能满足 guard_held 的前置条件；即使该准备姿态接近护架，也不提前计入此节点。成功参考录像在其视频约 0.24 s 后出现击打后的收拳保持，本次视频没有相应后续过程；参考录像仅用于辨认可见动作。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N*s",
+   "threshold": 12.0,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.1,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "hit_height_m",
+   "unit": "m",
+   "threshold": [
+    1.2,
+    1.6
+   ],
+   "direction": "target",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bag_struck",
+   "t": 2.24,
+   "log_step": 150,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 2.16–2.24 s 机器人右手前伸至白带袋面附近，身体仍站立；log step 150 对最后一局 episode 6 返回非右拳拳峰部位触袋的现场裁判判词，stages_reached=0/2。失败发生在这次击袋操作中，未完成有效击打，视频内也没有修复；视频 2.24 s 为末尾接触附近的粗略导航锚点。"
+  },
+  "unmet": {
+   "bag_struck": false,
+   "guard_held": false
+  },
+  "phenomena": [
+   {
+    "cat": "约束",
+    "item": "违反约束",
+    "t": 2.24,
+    "log_step": 150,
+    "evidence_zh": "观察：视频 2.16–2.24 s 右手末端到达沙袋白带附近；log step 150 中现场裁判明确返回“the bag was touched by something other than the knuckle part of the right fist”，随即结束该局，违反 wrong_bag_contact 所述只允许右拳拳峰触袋的规则。画面未见跌倒或被固定障碍阻挡，不另推断冲量不足、持续推袋或具体违规接触部位。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "抓取和接触设计",
+    "alt_feasible": "已验证有",
+    "log_step": 148,
+    "evidence_zh": "推断：最终击袋姿态的接触几何未能只让拳峰触袋。log step 148 的最终脚本把右腕偏航设在约 +1.27 rad，以侧向拳垫边缘先行；log step 149 的本地模型测试对最终采用的 acc=5、ex=1 仍报告拳垫接触后约 0.015 s 出现 right_finger_index 接触，log step 150 随后在现场得到非拳峰触袋判词。该模型是 agent 自建近似模型，不能证明真实接触的是哪根手指，但与视频末尾右手触袋及裁判结果共同支持接触布局不合适。可行替代仅由 inputs/oracle/oracle.json 的 success=1 和参考录像中其视频约 0.24 s 后收拳保持验证；不把参考动作计入本次运行。",
+    "source_zh": "日志"
+   }
+  ],
+  "gap_zh": "没有状态日志；视频仅有场馆侧视，手部像素不足以区分拳垫拳峰、侧缘与手指的具体接触，也不能独立测量冲量、接触历时或触点高度。inputs/run.json 的评分字段未报告 impulse_n_s、contact_s、hit_height_m，故三项原生指标均留 null；本地测试的约 14.9 N·s 不是现场裁判测量。agent 在 log step 150、152 所述手指突出尺寸和碰撞机理仍待核实，根因仅归到接触设计，不据此认定任务无合法解。"
+ },
+ "summary_zh": "受评的是最后一局 episode 6：机器人站立调整右臂后向沙袋白带出拳，视频约 2.24 s 的接触附近因非右拳拳峰部位触袋被现场裁判终止，有效击打和击后护架均未达成（0/2）。最终侧向出拳的接触布局未能避开违规触袋，与自建模型中仍出现手指接触的测试相符。缺少状态日志和现场冲量等数值，具体违规接触部位及细小几何尺寸无法独立核实。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "受评的最后一局 episode 6（log control_steps=226，约 2.26 s，与视频 2.28 s 吻合）：视频中右手贴近沙袋白带出拳，现场裁判返回非右拳拳峰部位触袋、stages_reached=0/2。bag_struck 与 guard_held 均为 false、断点 bag_struck/操作中、违反约束、接触设计根因均有 log 与评分字段支持；未发现需要改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:48:55Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--vwthikh/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-boxing-i00-standard--vwthikh/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "03ad0022b2ff24f84e2ffc57f1aad263c39e7722a62d13a553b5936bc8023cc9",
+   "video_sha256": "b23ee25775893e00fd1103b51c8c8470d4abfd35cad9215bbd3fa4ff8dd7060f",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--63uqpxj.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--63uqpxj.json
new file mode 100644
index 0000000000000000000000000000000000000000..46d0566768f5d5491986d581997ed718471ec11d
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--63uqpxj.json
@@ -0,0 +1,239 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-privileged--63uqpxj",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 0,
+  "reward": 0.5,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 1038,
+   "sim_steps": 1038,
+   "horizon": 3000,
+   "episode_over": 0,
+   "progress": 0.5,
+   "stages_reached": 1,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 8.71,
+   "bad_trajectory": 0,
+   "metric_ball_speed_m_s": 0.004,
+   "metric_ball_to_hole_m": 2.949,
+   "metric_grip_break_max_s": 0.004,
+   "metric_head_contact_s": 0.012,
+   "metric_head_hits": 1.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 139,
+   "agent_resets": 21,
+   "agent_replays": 3,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.943,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.981,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-privileged--63uqpxj/replay.mp4",
+  "duration_s": 10.4,
+  "fps": 25.0,
+  "frames": 260,
+  "sha256": "6fb3cb25bd57fecf88b70db14b09aef02aeda08794c70123aa9294967589f368"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00-9.40 s，机器人双脚着地，右手持推杆缓慢转动躯干并调整杆头，白球仍在脚前，球洞位于画面右上方浅色条带上。"
+     },
+     {
+      "t": 9.52,
+      "log_step": null,
+      "text_zh": "观察：视频 9.44-10.36 s，机器人挥动推杆后白球向画面右侧（略偏下）短距离滚动，仍留在机器人脚前，未向右上方洞口滚去，更未落入洞中；已逐帧检查视频 9.28-10.36 s。"
+     },
+     {
+      "t": 10.36,
+      "log_step": 212,
+      "text_zh": "观察：log step 212 确认最终保存并重放的是 1038 行轨迹，仅有 stroke 阶段，球距洞 2.949 米；inputs/run.json 评分器字段同样为 stages_reached=1/2、metric_head_hits=1、metric_head_contact_s=0.012、metric_ball_to_hole_m=2.949、success=0。视频 10.36 s 白球仍在脚前，与该结果一致。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 9.5 s，机器人右手挥动所持推杆，杆头经过白球后白球开始滚动；评分器记录杆头触球 1 次、累计 0.012 秒。候选接触名为“击打”，现有接触方式无击打项；按工具至白球链路记录，不重复标握柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00-10.36 s，机器人双脚在果岭上支撑身体，转动躯干并用右手调整和挥动推杆；击球及后续球滚动时未见跌倒。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": 2.949
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": 0.004
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": 0.012
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.004
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "ball_holed",
+   "t": 9.5,
+   "log_step": null,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 9.5 s，机器人挥杆触球后，白球向画面右侧（略偏下）短距离滚动，偏离通往右上方球洞的路线，仍在脚前；视频 10.36 s 仍未入洞。评分器记录一次杆头触球而仅达到 stroke 阶段，末态球距洞 2.949 米；失败发生在一杆推球入洞的击球操作中。"
+  },
+  "unmet": {
+   "ball_holed": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 9.5,
+    "log_step": null,
+    "evidence_zh": "观察：视频 9.44-10.36 s，机器人挥杆后白球向画面右侧（略偏下）短距离滚动，球的路线没有朝向画面右上方浅色条带上的球洞；至视频 10.36 s，球仍在机器人附近。评分器记录末态球距洞 2.949 米，未完成入洞。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "source_zh": "日志",
+    "evidence_zh": "推断：log step 209 的交付脚本先调整右手及杆面朝向，再命令右手沿 +y 挥动；会话中同配置试验的输出显示击球准备时杆面中心 x=0.3544 米，而初始球心 x=0.44 米，击出后的球速主要沿 +x（约 [0.55,0.14,0] 米/秒），而球洞在 +y 方向。log step 211-212 将该配置生成并保存为最终 1038 行轨迹，重放仅达到 stroke、球距洞 2.949 米；视频 9.44-10.36 s 同样显示侧向短滚。因此可归因于击球对位及挥杆控制未能产生朝洞的有效击球，具体控制子类证据不足。"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志，远景无法独立核实杆头具体哪一部位触球，也无法核实 agent 在 log step 251 自述的关节饱和、承载下沉及杆头趾部擦球机制，因此根因停在“控制”。视频为 25 fps，毫秒级接触时长取自评分器字段，断点仅锚定视频约 9.5 s。"
+ },
+ "summary_zh": "机器人完成了持杆调整及一次触球，但白球只在脚前侧向短滚，未入洞，唯一语义节点最终为 0/1。断点位于视频约 9.5 s 的击球操作，现象为位姿不准；交付脚本及同配置试验支持击球对位与挥杆控制不足，评分器末态球距洞 2.949 米。缺少逐帧状态和杆头近景，不能进一步确认具体接触或跟踪机制；会话后续滚得更远的测试版本不属于最终交付。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 7,
+    "summary_zh": "节点未达成(final=false)、断点 ball_holed/操作中、位姿不准加控制根因与评分器字段及帧一致；仅球滚动方向的描述与视频不符，需小改。",
+    "n_issues": 1,
+    "n_major": 0
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点未达成、断点 ball_holed/操作中、位姿不准与评分器字段（stages 1/2、球距洞 2.949 m）及视频一致；上一轮方向描述已修正，未发现新错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:59:45Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-privileged--63uqpxj/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-privileged--63uqpxj/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-privileged--63uqpxj/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-privileged--63uqpxj/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "357f10b2224cc60fb23f4cad56cfe07a0b838b72bebff42ee7c488f6f3db917e",
+   "video_sha256": "6fb3cb25bd57fecf88b70db14b09aef02aeda08794c70123aa9294967589f368",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--btzgldk.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--btzgldk.json
new file mode 100644
index 0000000000000000000000000000000000000000..c59767e1ef3537ae341d818e0c11e62e378e7d79
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--btzgldk.json
@@ -0,0 +1,201 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-privileged--btzgldk",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 664,
+   "sim_steps": 664,
+   "horizon": 3000,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 5.28,
+   "bad_trajectory": 0,
+   "metric_ball_speed_m_s": 0.0,
+   "metric_ball_to_hole_m": 0.002,
+   "metric_grip_break_max_s": 0.007,
+   "metric_head_contact_s": 0.025,
+   "metric_head_hits": 2.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 961,
+   "agent_resets": 260,
+   "agent_replays": 52,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-privileged--btzgldk/replay.mp4",
+  "duration_s": 6.68,
+  "fps": 25.0,
+  "frames": 167,
+  "sha256": "9ef647976e496b781f1cf3ebf11162adafece52ebe6a2710f05507711fd49433"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.64,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00 s，白球位于机器人脚前，远处浅色条带上可见洞口；机器人右手已持推杆，双脚落在果岭上，入洞条件尚未成立。"
+     },
+     {
+      "t": 2.36,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 2.08–2.68 s，机器人右臂带动推杆挥过脚前白球，随后白球离开原位向画面右上方的球洞运动；视频 2.60–4.48 s 可连续追踪白球沿果岭接近洞口，机器人双脚保持支撑。"
+     },
+     {
+      "t": 4.96,
+      "log_step": null,
+      "text_zh": "观察：视频 4.48–4.96 s 的洞口放大序列中，白球从洞口近侧滚至洞沿后消失；视频 5.60–6.64 s 的密集末段帧中，洞口及附近果岭不再露出白球，机器人仍站立持杆。远景看不到洞内球体，不能直接量出球顶高度或毫米级静止。"
+     },
+     {
+      "t": 6.64,
+      "log_step": 330,
+      "text_zh": "观察：log step 329 重新运行交付文件的生成脚本；log step 330 对最终 trajectory.npz 的两次独立重放均由仿真裁判返回 success=true、failed=null、stages=[stroke, holed]，执行 664 个控制步，time_s=6.64，末态哈希一致。两次事件记录均为仿真 2.358 s 的 stroke、仿真 6.64 s 的 holed；run.json 的评分器字段 success=1、replay_success=1、deterministic=1 与之相符。"
+     },
+     {
+      "t": 6.64,
+      "log_step": 330,
+      "text_zh": "推断：log step 330 中最终交付的 664 控制步对应仿真 6.64 s，即每步 0.01 s；视频共 167 帧、25 fps，末帧为视频 6.64 s（封装时长 6.68 s）。重放的仿真 2.358 s 击球事件与视频约 2.36–2.60 s 的挥杆及白球离开原位相符，据此将末次 holed 验收定位到视频 6.64 s；结合白球在洞口消失并一直未再出现，节点自该时刻成立且终值为真。洞内停稳 1.0 秒及细粒度接触合规由最终重放裁判确认，并非仅凭白球消失提前记为完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 2.36–2.68 s，机器人右手持推杆挥过白球，白球随后离开脚前并滚向球洞。作用端为推杆杆头、对象为白球；候选接触名为“击打”，现有方式无击打项，因此用“其他”；握杆属于同一工具链，不重复计为抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–6.64 s，机器人双脚接触果岭支撑身体，挥杆击球及随后等待期间均保持站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": 0.002
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": 0.0
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": 0.025
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.007
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人完成“一杆推球入洞”，终态为 1/1：视频约 2.36–2.60 s 右手持杆击球，白球约在视频 4.9 s 到洞口后消失，机器人保持站立持杆至结束。最终交付的两次重放在 log step 330 均记录 stroke 后 holed、成功且末态哈希一致，完整停稳验收对应视频 6.64 s，未发现失败断点。无状态日志，远景无法直接核验洞内毫米级静止、球顶高度和毫秒级接触，相关判据采用最终重放裁判及评分器结果；没有另用独立成功参考视频。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "成功运行，单节点 ball_holed 终值为真；时钟映射（664 步×0.01 s=6.64 s）、stroke 2.358 s 与视频挥杆相符，接触标签合理，无失败断点，未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:45:04Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-privileged--btzgldk/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-privileged--btzgldk/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "2f804a776c08be9ce2182374a377b3a057b0376015bd42f75f8cb9c80b2ce8ab",
+   "video_sha256": "9ef647976e496b781f1cf3ebf11162adafece52ebe6a2710f05507711fd49433",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--cbn3ffv.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--cbn3ffv.json
new file mode 100644
index 0000000000000000000000000000000000000000..c71aa4f9caa395157f7bbfbec8a174ad904cd0b0
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--cbn3ffv.json
@@ -0,0 +1,198 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-privileged--cbn3ffv",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 782,
+   "sim_steps": 782,
+   "horizon": 3000,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 7.39,
+   "bad_trajectory": 0,
+   "metric_ball_speed_m_s": 0.003,
+   "metric_ball_to_hole_m": 0.026,
+   "metric_grip_break_max_s": 0.001,
+   "metric_head_contact_s": 0.013,
+   "metric_head_hits": 1.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 72,
+   "agent_resets": 56,
+   "agent_replays": 5,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.664,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.242,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-privileged--cbn3ffv/replay.mp4",
+  "duration_s": 7.84,
+  "fps": 25.0,
+  "frames": 196,
+  "sha256": "91413cc97a9c432b42b89a3231a0671ba73a4a20b55cc78eb33c06b9eca8649e"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.8,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.12,
+      "log_step": null,
+      "text_zh": "观察：视频 3.76–4.08 s，机器人右手持推杆将杆头挥向脚前白球；视频 4.12–4.96 s，白球离开原位并沿果岭向浅色条带上的洞口滚动，机器人双脚持续着地支撑。"
+     },
+     {
+      "t": 6.32,
+      "log_step": null,
+      "text_zh": "观察：视频 5.84–6.16 s 白球接近洞沿，视频约 6.2–6.4 s 白球在洞口处落下；视频 6.4–7.80 s 洞外及洞沿没有可辨识的露出白球，机器人仍站立持杆。远景无法辨明球在洞内的毫米级运动。"
+     },
+     {
+      "t": 7.8,
+      "log_step": 74,
+      "text_zh": "观察：inputs/run.json 的评分器字段 success=1、replay_success=1、deterministic=1、stages_reached=2/2、metric_head_hits=1、fell=0；log step 74 中 agent 重新生成最终交付文件后，两次官方重放均返回 success 且末态相同，阶段为 stroke、holed，指标与最终评分一致。"
+     },
+     {
+      "t": 7.8,
+      "log_step": 74,
+      "text_zh": "推断：结合视频约 6.3 s 落洞及视频 7.80 s 末帧仍无洞外白球、最终评分成功和 log step 74 的最终文件重放结果，将包含洞内停稳 1.0 秒的完整节点标为在视频约 7.8 s 达成并保持至结束；这是验收完成的近似末帧锚点，远景不能确定毫米级停稳窗口的精确起点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 4.08–4.16 s，机器人右手持推杆挥过白球原位，白球随即滚离；评分器 metric_head_hits=1、metric_head_contact_s=0.013 支持杆头击球。候选接触名为“击打”，现有方式无对应条目；仅记录推杆杆头至白球的工具链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0–7.80 s，机器人双脚着地，原地站立完成调杆、击球和等待白球入洞，未见跌倒。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": 0.026
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": 0.003
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": 0.013
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.001
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人右手持杆、双脚原地支撑，在视频约 4.1 s 击出白球，白球沿果岭滚动并于视频约 6.3 s 落入洞口；完整的一杆入洞及停稳节点在视频约 7.8 s 标为完成，最终达到 1/1。评分器判定成功，最终交付文件的两次官方重放结果一致，单次杆头击球累计接触 0.013 秒。没有状态日志，远景无法独立核验全球低于表面及毫米级停稳窗口，精细验收依据评分器与最终文件重放结果。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "成功运行，单节点 ball_holed 达成、final=true，与评分器 success=1、stages 2/2、metric_ball_to_hole 0.026 一致；远景下击球与滚入洞口的描述与帧相符，接触标注（手持工具/其他、腿/躯干/支撑）合理，无失败层。7.8 s 为近似锚点，证据已明示其为推断。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:44:50Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-privileged--cbn3ffv/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-privileged--cbn3ffv/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "5bf0706f638da880ac0ceb71545099a54b5870247fc4080e306f68b9fcc054de",
+   "video_sha256": "91413cc97a9c432b42b89a3231a0671ba73a4a20b55cc78eb33c06b9eca8649e",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--hvt294e.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--hvt294e.json
new file mode 100644
index 0000000000000000000000000000000000000000..2ae964379f4ddee0998f8fbcc2615d273ed9a93b
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--hvt294e.json
@@ -0,0 +1,203 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-privileged--hvt294e",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 413,
+   "sim_steps": 413,
+   "horizon": 3000,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 4.89,
+   "bad_trajectory": 0,
+   "metric_ball_speed_m_s": 0.0,
+   "metric_ball_to_hole_m": 0.022,
+   "metric_grip_break_max_s": 0.001,
+   "metric_head_contact_s": 0.012,
+   "metric_head_hits": 1.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 86,
+   "agent_resets": 309,
+   "agent_replays": 5,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.743,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.253,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-privileged--hvt294e/replay.mp4",
+  "duration_s": 4.16,
+  "fps": 25.0,
+  "frames": 104,
+  "sha256": "d0e9341a8275b37152baaca7a59bdfbc3783d68f6e675230b83154d54296bbbd"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.12,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.96,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.92 s，机器人双脚支撑于果岭，右手持推杆调整并回摆；视频 0.96–1.04 s 杆头向前扫过白球原位，白球离开脚前位置，随后沿果岭朝远处浅色条带上的洞口滚动。"
+     },
+     {
+      "t": 2.6,
+      "log_step": null,
+      "text_zh": "观察：视频 2.24–2.60 s，白球逐渐接近圆形洞口，随后在洞口处不再可辨；逐帧检查视频 3.12–4.12 s，洞口及洞沿均未再见白球，机器人继续站立持杆。成功参考视频使用相同远景视角，入洞后也呈现洞口无白球的外观；参考只用于辨认视觉线索。"
+     },
+     {
+      "t": 4.12,
+      "log_step": null,
+      "text_zh": "观察：inputs/run.json 的评分器 reward 记录 success=1、replay_success=1、deterministic=1、stages_reached=2/2、episode_over=1，末态球距洞心 0.022 m、球速 0 m/s，杆头触球累计 0.012 s、head_hits=1。"
+     },
+     {
+      "t": 4.12,
+      "log_step": null,
+      "text_zh": "推断：视频中的击球、滚向洞口并消失与评分器成功结果一致，故一杆入洞且洞内停稳验收成立，final=true。远景不能测量全球高度、毫米级位移或洞内停稳计时起点，达成时间保守取视频 4.12 s 末帧作近似锚点，精确首次成立时间未知；未把 agent 工作时间或重放耗时换算为视频时间。"
+     },
+     {
+      "t": null,
+      "log_step": 153,
+      "text_zh": "观察：log step 147 的脚本生成并保存击球轨迹；log step 153 写出的最终生成器包含准备、回摆、前挥及等待回合结束四段，log step 154 执行生成器，log step 157 发起交付文件重放检查，log step 163 读出交付轨迹为 413 行有限动作。agent 在 log step 166 自述已一杆入洞；节点结论依据本次视频和评分器结果，未把会话调试回合当作评分回合。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 0.92–1.04 s，机器人右手持推杆前挥，杆头经过白球原位后白球向洞口运动；评分器记录一次杆头接触。候选接触名为“击打”，现有方式无对应项；只记录杆头至白球的工具接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–4.12 s，机器人双脚接触果岭承重，挥杆及等待球滚动时持续站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": 0.022
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": 0.0
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": 0.012
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.001
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人右手持推杆击出白球，白球沿果岭滚至洞口并消失，评分器确认一杆入洞成功，唯一节点完成，终态为 1/1。视频 3.12–4.12 s 洞口及洞沿未再见白球，评分器同时记录确定性重放成功、末态球速为零；无失败断点。没有状态日志，远景无法精确复核洞内高度、毫米级停稳及毫秒级接触，节点达成时间以视频 4.12 s 末帧为保守近似锚点。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "单节点 ball_holed 成功，视频 2.5 s 左右白球入洞消失，之后洞口无球，评分器 success=1；4.12 s 作为含 1 s 停稳窗口的达成锚点与我的判断（约 3.5 s）相差不到一秒，接触标注与 DAG 一致且符合画面，无失败断点。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:44:53Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-privileged--hvt294e/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-privileged--hvt294e/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "63993beb9638d28408d2acc3895e64aaaa0f925c41108f931c965c654c3e627d",
+   "video_sha256": "d0e9341a8275b37152baaca7a59bdfbc3783d68f6e675230b83154d54296bbbd",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--uawrqwe.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--uawrqwe.json
new file mode 100644
index 0000000000000000000000000000000000000000..7eded543d44282ef66fff015c828782cc046548d
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-privileged--uawrqwe.json
@@ -0,0 +1,198 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-privileged--uawrqwe",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 597,
+   "sim_steps": 597,
+   "horizon": 3000,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 9.49,
+   "bad_trajectory": 0,
+   "metric_ball_speed_m_s": 0.0,
+   "metric_ball_to_hole_m": 0.025,
+   "metric_grip_break_max_s": 0.003,
+   "metric_head_contact_s": 0.013,
+   "metric_head_hits": 1.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 34,
+   "agent_resets": 29,
+   "agent_replays": 4,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.702,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.251,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-privileged--uawrqwe/replay.mp4",
+  "duration_s": 6.0,
+  "fps": 25.0,
+  "frames": 150,
+  "sha256": "a770514f7c3c43f2242e6806b24252395211af938e1fca3a493387609cfa0471"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.96,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.4,
+      "log_step": null,
+      "text_zh": "观察：视频 2.24–2.68 s 的逐帧画面中，机器人双脚站在果岭上，右手持推杆向白球挥动；视频 2.40 s 白球仍在杆头前，视频 2.44–2.68 s 白球离开原位并向远处洞口滚动，杆头留在机器人附近。"
+     },
+     {
+      "t": 4.2,
+      "log_step": null,
+      "text_zh": "观察：视频 3.20–4.16 s 白球持续靠近浅色条带上的圆形洞口；视频约 4.20–4.24 s 白球在洞口处消失。逐帧检查视频 4.40–5.96 s，洞口及洞沿未见白球重新露出；末帧机器人仍站立持杆。参考成功视频的末尾同样只见洞口，远景均不能显示洞底细节。"
+     },
+     {
+      "t": 5.96,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器 reward 记录 success=1、replay_success=1、deterministic=1、stages_reached=2/2，终态 ball_to_hole_m=0.025、ball_speed_m_s=0，结合白球在洞口消失且未再露出的画面，判定入洞停稳验收已通过。视频不能精确显示洞内静止窗口的起止，故将达成时间近似锚定于已检查的末帧视频 5.96 s（约视频 6.0 s），不把白球初入洞时刻当作完成停稳。"
+     },
+     {
+      "t": null,
+      "log_step": 73,
+      "text_zh": "观察：log step 59 修正生成脚本的保存调用，log step 60 调用修正后的脚本生成并重放交付文件；log step 73 显示最终脚本仍采用抬杆中间姿态、击球前停留及单次挥杆方案。log step 76 中 agent 自述交付含 597 个控制步并已独立重放；这些会话记录用于核对交付内容，节点值依据本次视频和评分器结果。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.40–2.68 s，机器人右手握持的推杆杆头击出白球，白球随即滚离；候选接触名为“击打”，现有接触方式没有对应项，短促击球不属推/扫。评分器记录 head_hits=1、head_contact_s=0.013 秒；工具链仅记录杆头与白球，不重复记录握杆。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–5.96 s，机器人双脚在果岭上支撑身体，右臂完成移杆、挥杆及随挥，击球和等待白球入洞期间均保持站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": 0.025
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": 0.0
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": 0.013
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.003
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人右手持推杆约在视频 2.4 s 击出白球，白球约在视频 4.2 s 于洞口消失，至视频 5.96 s 末帧未再露出；评分器确认入洞停稳成功，唯一节点最终完成。节点完成时间近似标于视频末尾；没有状态日志，远景无法独立精确复核全球高度、毫米级静止窗口及毫秒级接触，相关验收由本次交付的评分器结果补足。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "成功运行，唯一节点 ball_holed 终值 true；视频 4.24 s 球在洞口消失、末帧未再露出，与评分器结果一致。时间锚点 5.96 s 略晚于洞内停稳约 5.2 s，但在容差内；接触标注合理。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:44:50Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-privileged--uawrqwe/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-privileged--uawrqwe/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "6b4652d4b4296feaf7a9d117baebb77c55e3fc68d3fedaf1de1ea17c5c9ffc45",
+   "video_sha256": "a770514f7c3c43f2242e6806b24252395211af938e1fca3a493387609cfa0471",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--ac3ubxs.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--ac3ubxs.json
new file mode 100644
index 0000000000000000000000000000000000000000..aef5f1b446bd224bbbb3099ba676999639e009d1
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--ac3ubxs.json
@@ -0,0 +1,222 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-standard--ac3ubxs",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 1157,
+   "sim_steps": 3341,
+   "calls": 35,
+   "rows": 1501,
+   "resets": 3,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 7064.3,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.304,
+   "peak_processes": 12,
+   "peak_disk_gb": 2.542,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-standard--ac3ubxs/replay.mp4",
+  "duration_s": 11.6,
+  "fps": 25.0,
+  "frames": 290,
+  "sha256": "740b20fe0bc3fa7d36ed9a369b7066e524438ac1eb1a4c03954cb687e76fa5eb"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00 s，机器人右手持推杆站在果岭上，白球在脚前，球洞位于远处浅色条带上，初始未入洞。"
+     },
+     {
+      "t": 3.8,
+      "log_step": 62,
+      "text_zh": "观察：视频 2.80–4.40 s 的逐帧画面中，机器人转动腰部、右手带杆先回摆再前挥，白球始终留在脚前，没有随挥杆滚向球洞。log step 62 的服务记录为 episode=4，提交 557 个控制步后又观察 600 个控制步，总计 1157 步，现场尚未成功。"
+     },
+     {
+      "t": 11.56,
+      "log_step": 64,
+      "text_zh": "观察：视频 10.60–11.56 s 的末秒画面及末帧，白球仍在脚前原处，机器人持杆站立，未出现白球滚到洞口或落洞的过程。log step 64 中 agent 自述“Ball never moved — a whiff”，与画面一致。"
+     },
+     {
+      "t": 5.6,
+      "log_step": 62,
+      "text_zh": "推断：log step 3 给出控制周期 0.01 s，log step 62 的本回合 1157 个控制步对应 11.57 s，与视频 11.60 s 长度吻合；557 步对应视频约 5.57 s，已用视频 5.60 s 挥杆结束后的持杆静止姿态核对。该视频对应最后的 episode=4；log step 54 中 episode=3 的击球及洞外停球属于重置前测试，不计入本次进度。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.00–11.56 s，机器人右手围握推杆握柄并带杆挥动；本次杆头未击中白球，因此不填写手持工具击球接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–11.56 s，机器人双脚接触果岭支撑身体，腰部转动挥杆期间保持站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "ball_holed",
+   "t": 3.8,
+   "log_step": 62,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 3.60–4.00 s，机器人右手持推杆完成前挥，但杆头从白球旁经过，白球没有移动；至视频 11.56 s 仍未入洞，后续没有补救动作。log step 62 是最后一次重置后的现场回合，log step 54 的前一回合击球结果不属于本次。"
+  },
+  "unmet": {
+   "ball_holed": false
+  },
+  "phenomena": [
+   {
+    "cat": "接触",
+    "item": "没建立接触",
+    "t": 3.8,
+    "log_step": 62,
+    "evidence_zh": "观察：视频 2.80–4.40 s 逐帧可见推杆回摆和前挥，白球保持在脚前同一位置；视频 10.60–11.56 s 仍在原处。推断：这次挥杆未建立将白球击出的杆头接触，而非击球后未能入洞。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 3.8,
+    "log_step": 64,
+    "source_zh": "日志",
+    "evidence_zh": "推断：log step 61 计算对准球后的右臂目标，log step 62 将该目标用于实际挥杆；log step 64 读取这次提交保存的本体感知，腰部、右肘及右肩的采样实际角度接近对应指令，但视频 3.60–4.00 s 的实际杆头仍未触球，说明挥杆控制没有使实际球杆路径与球匹配。agent 在 log step 65 将其解释为握柄内滑动，此解释没有独立接触或球杆姿态数据验证，故根因仅标控制，不细分。"
+   }
+  ],
+  "gap_zh": "没有状态日志；远景画面能确认挥杆后白球不动，但不能精确区分杆头偏离是由握柄内转动、实际球杆几何与局部模型不一致，还是其他控制误差引起。agent 在 log step 65 所称的滑动尚未独立核实；评分器未报告 DAG 的四项原生指标，均填 null。"
+ },
+ "summary_zh": "本次评分视频对应最后一次现场回合 episode=4：机器人保持站立、右手持杆完成回摆和前挥，但白球始终留在脚前，一杆推球入洞节点未达，最终为 0/1。断点在视频约 3.80 s 的挥杆操作，现象为没建立接触，根因归控制；远景及缺少状态日志使具体杆头偏离机制无法核实。日志中此前重置回合的击球不计入本次。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标签与视频、log 一致：视频 11.6 s 对应 episode 4 的 1157 个控制步（0.01 s/步），白球全程留在原处，ball_holed 未达，断点在挥杆，现象没建立接触、根因控制（不细分）均有证据且已声明不确定性。未发现需修改的问题。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:00:27Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--ac3ubxs/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--ac3ubxs/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "4e8bbf4c093d304a867ee2cb1d63a29954b9aa4eb998fd09593a502b83bc147d",
+   "video_sha256": "740b20fe0bc3fa7d36ed9a369b7066e524438ac1eb1a4c03954cb687e76fa5eb",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--ekj9qc2.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--ekj9qc2.json
new file mode 100644
index 0000000000000000000000000000000000000000..387f9a951d6c8cd0e6c3d9b02075111ef3ba41b6
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--ekj9qc2.json
@@ -0,0 +1,191 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-standard--ekj9qc2",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "live_success": 1,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 647,
+   "sim_steps": 1409,
+   "calls": 6,
+   "rows": 1409,
+   "resets": 1,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 3154.1,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.895,
+   "peak_processes": 7,
+   "peak_disk_gb": 0.361,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-standard--ekj9qc2/replay.mp4",
+  "duration_s": 6.48,
+  "fps": 25.0,
+  "frames": 162,
+  "sha256": "36afd362c723d0208d5f1d935c30f420fb8ee262bea94939c65088a26592eb5a"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.47,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.8,
+      "log_step": null,
+      "text_zh": "观察：视频 0.0–2.6 s，机器人双脚站在果岭上，右手持推杆调整至白球后方；视频 2.72–2.84 s，右臂挥杆，杆头经过白球，白球离开脚前原位并朝远处浅色条带上的洞口滚动。"
+     },
+     {
+      "t": 5.0,
+      "log_step": null,
+      "text_zh": "观察：视频 4.20–4.92 s 的逐帧洞口裁剪中，白球从洞口近侧接近并与洞口重合；视频约 5.0 s 后球不再露出，直至末帧视频 6.44 s 未见球重新出现在洞沿或果岭。远景能显示滚入洞口的过程，不能分辨洞内球顶高度与毫米级位移。"
+     },
+     {
+      "t": 6.47,
+      "log_step": 130,
+      "text_zh": "推断：log step 126–127 执行重置后的第二回合，log step 130 的机器人服务末态为 episode=2、control_steps=647、done=true、reason=success，评分器 reward 的 live_success=1、graded_by_live=1。log step 3/6 给出 control_dt=0.01 s，因此本回合终止对应视频约 647×0.01=6.47 s，与视频总长 6.48 s 一致；log step 94/124 的脚本在本回合控制时钟 2.5 s 后挥杆，已与视频 2.6–2.84 s 的实际挥杆核对。结合入洞后不再露出的画面，以视频约 6.47 s 作为含洞内停稳一秒在内的首次完成时刻；不是把 agent 工作时间换算为视频时间。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.72–2.84 s，机器人右手所持推杆的杆头经过白球并将球击出，随后球独立滚向洞口；候选接触名为“击打”，现有方式无对应项。按工具至球的链路记录，不另列握杆接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.0–6.44 s，机器人双脚支撑于果岭，站立调整右臂、挥杆并保持随挥姿态，未见跌倒。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": null,
+ "summary_zh": "受评的是重置后的第二回合：机器人右手持杆击出白球，白球滚入洞口后未再露出，服务与评分器均判成功，唯一节点“一杆推球入洞”在视频约 6.47 s 完成。第一回合未进洞的调试结果不计入本次进度；本次无失败断点。没有状态日志，远景无法独立核验洞内毫米级停稳、球顶高度及毫秒级接触阈值，这些细节依据本回合成功裁决；四项原生指标未在评分器字段报告，均记 null。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "成功回合，唯一节点 ball_holed 在视频约 6.47 s 完成，与视频、控制步数 647×0.01 s 及评分器一致；接触与无失败断点均合理。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:45:42Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--ekj9qc2/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--ekj9qc2/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "e2844cbc9479f5e90a9ab96e0ad6b7383b1410a5a1bb2605a5b7ac22441b2891",
+   "video_sha256": "36afd362c723d0208d5f1d935c30f420fb8ee262bea94939c65088a26592eb5a",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--he4xdrj.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--he4xdrj.json
new file mode 100644
index 0000000000000000000000000000000000000000..98f378c742b313f0f46ec48b7d8928320bf553e2
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--he4xdrj.json
@@ -0,0 +1,238 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-standard--he4xdrj",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 1503,
+   "sim_steps": 13580,
+   "calls": 55,
+   "rows": 13580,
+   "resets": 14,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 5149.6,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.296,
+   "peak_processes": 10,
+   "peak_disk_gb": 1.955,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-standard--he4xdrj/replay.mp4",
+  "duration_s": 15.04,
+  "fps": 25.0,
+  "frames": 376,
+  "sha256": "1daade181cb7928775e7a6aa5d0a02fcd3509b2885d1f6439704cb9f0ad5408f"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00 s 白球在机器人脚前，球洞在远处浅色条带上；机器人右手已持推杆。"
+     },
+     {
+      "t": 6.4,
+      "log_step": null,
+      "text_zh": "观察：视频 5.60–8.20 s 机器人右手带动推杆摆动，杆头在白球的图像左侧（起点侧）摆动，未到达白球，白球未见位移；视频 8.40–8.88 s 收杆时杆头投影靠近白球后移至其图像右侧，白球仍留在原处；视频 12.00–15.00 s 球杆垂在手下，白球仍未滚向球洞。"
+     },
+     {
+      "t": 15.0,
+      "log_step": 312,
+      "text_zh": "观察：视频 14.00–15.00 s 白球仍在原起点附近，远离洞口，未出现滚向洞口并落下的过程。log step 312 的末次现场回合输出为 OUTCOME: None、control_steps=1503；评分字段 live_success=0、graded_by_live=1 与未入洞画面一致。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.00–15.00 s 机器人右手握持推杆并带动它摆动；这里记录实际持杆接触，未确认杆头对白球形成击打，不把计划中的工具击球接触记为实际发生。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–15.00 s 机器人以双脚在果岭上支撑，挥杆及恢复姿态时有踏步和身体位移。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "ball_holed",
+   "t": 6.4,
+   "log_step": 312,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 5.60–8.20 s 机器人已持杆挥动，但杆头在白球的图像左侧摆动，未到达白球，白球没有被送出；视频 8.40–8.88 s 收杆时杆头投影靠近球后移至图像右侧，白球仍未见位移，视频 15.00 s 仍在起点附近。log step 312 是最后一次重置后的现场尝试，输出 OUTCOME: None steps 1503；reward 的 graded_by_live=1、n_actions=1503、live_success=0 表明评分取此末次回合，较早试验的推球、非法触球和跌倒结果不用于本回合现象。"
+  },
+  "unmet": {
+   "ball_holed": false
+  },
+  "phenomena": [
+   {
+    "cat": "接触",
+    "item": "没建立接触",
+    "t": 6.4,
+    "log_step": null,
+    "evidence_zh": "推断：视频 5.60–8.20 s 机器人右手挥动时杆头在白球的图像左侧摆动，未到达球；视频 8.40–8.88 s 收杆时虽有杆头投影靠近球并移至其图像右侧，白球仍未见位移，视频 12.00–15.00 s 也仍在原处。据此判断本次挥杆未建立所需的有效杆头击球接触；远景不足以排除没有可见位移的极短擦碰。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "log_step": 312,
+    "t": 6.4,
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "推断：log step 303 的脚本以手部位置目标生成右臂逆解及恒加速度挥杆，log step 312 将手部高度目标从 0.885 m 降至 0.80 m 后执行末次回合；视频 5.60–8.20 s 机器人右手和推杆确实摆动，但杆头未到达白球，视频 15.00 s 白球仍在原处。该手部轨迹及其实际执行未使杆头到达所需击球位置，支持控制层归因；未记录实际手部及关节跟踪误差、杆头三维轨迹和接触诊断，不能区分手部路径与杆摆动设计、轨迹或逆解、反馈跟踪问题，故保留“控制”而不指定子类。是否存在可行替代：未知；不把 agent 自述的松摆机制当成已验证根因。",
+    "alt_feasible": "未知"
+   }
+  ],
+  "gap_zh": "没有状态日志；远景中球和杆头仅占数个像素，无法排除未引起可见位移的极短擦碰，也无法量出握持接触、杆头速度或接触时长。视频 8.40–8.88 s 收杆时杆头投影靠近球后移至其图像右侧，因此审查意见所称视频 5.60–12.00 s “始终在球左侧”不适用于整段；主挥杆未到达球与收杆投影经过球附近分开记录。末次回合服务输出未报告 stroke 或接触诊断，reward 也未报告 DAG 的四项原生指标，均保留 null；缺少实际关节跟踪及杆头三维轨迹，无法进一步区分手部路径与杆摆动设计、持杆柔顺性、轨迹或逆解和反馈控制，故根因仅标控制，可行替代为未知。"
+ },
+ "summary_zh": "本次评分回合中机器人右手持杆并进行了挥杆和姿态恢复，但白球始终留在起点附近，唯一节点“一杆推球入洞”未达到，最终为 0/1。断点在视频约 6.40 s 的挥杆：主要挥杆段杆头未到达球，后续收杆时杆头投影移过球附近也未使球发生可见位移，所需的有效击球接触未建立；较早重置回合的违规和跌倒不属于本次评分画面。根因保留控制层，因缺少状态、接触及跟踪诊断，无法确定具体机制或毫秒级擦碰，置信度维持 7。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 7,
+    "summary_zh": "节点未达成、断点 ball_holed、现象“没建立接触”与时钟均成立（视频 15.04 s = 1503 控制步，挥杆约 5.9 s 起与 log step 303/312 一致）。但证据把杆头描述为“从白球旁摆过/经过”，画面并不支持：杆头全程停在白球的起点侧（图中左侧），始终未到达球。需改写观察文字，并据此重新审视根因归类。",
+    "n_issues": 2,
+    "n_major": 1
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点 ball_holed 未达成、断点、现象“没建立接触”、根因“控制”（alt_feasible 未知）及时钟均有证据支持；上一轮意见已按画面修正，未发现新的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:00:58Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--he4xdrj/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--he4xdrj/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--he4xdrj/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--he4xdrj/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "f7ee66836cf6656de5a35385ae427c4a9f6dcb949a74ac4d143c15bbc71df201",
+   "video_sha256": "1daade181cb7928775e7a6aa5d0a02fcd3509b2885d1f6439704cb9f0ad5408f",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--jq97mv4.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--jq97mv4.json
new file mode 100644
index 0000000000000000000000000000000000000000..15fa03da563ad5e2fdb9748a8f51dac9851617e0
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--jq97mv4.json
@@ -0,0 +1,201 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-standard--jq97mv4",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "live_success": 1,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 1473,
+   "sim_steps": 3263,
+   "calls": 32,
+   "rows": 2575,
+   "resets": 2,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 1469.1,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.309,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.867,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-standard--jq97mv4/replay.mp4",
+  "duration_s": 14.76,
+  "fps": 25.0,
+  "frames": 369,
+  "sha256": "64b3466ad465cf5bf9380ddda4f93beac705d6a379ed8b0873bd2803c2bd902f"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [
+     [
+      14.7,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00 s 白球位于机器人脚前，球洞在远处浅色条带上，入洞条件尚未成立；机器人开局右手持杆，双脚站在果岭上。"
+     },
+     {
+      "t": 10.12,
+      "log_step": 96,
+      "text_zh": "观察：视频 9.60–10.00 s 机器人将杆头置于白球后方；视频 10.04–10.16 s 右手持杆向前挥动，杆头经过白球，视频 10.24–10.40 s 白球已离开原位并向洞口滚动。视频 10.60–12.32 s 白球继续接近球洞，机器人维持双脚支撑。"
+     },
+     {
+      "t": 13.2,
+      "log_step": null,
+      "text_zh": "观察：视频 12.32–13.12 s 白球到达洞口附近并在洞沿处短暂可见；视频 13.20–13.76 s 及视频 13.80–14.72 s 的密集帧中，洞口及洞沿已不再露出白球，机器人仍站立持杆。远景无法直接分辨洞内球心的毫米级运动。"
+     },
+     {
+      "t": 14.72,
+      "log_step": 102,
+      "text_zh": "观察：log step 102 的机器人现场服务返回 done=True、reason=success、outcome=success、stages_reached=2、control_steps=1473，传感读数 holding_right=1.0；inputs/run.json 的评分器字段 live_success=1、graded_by_live=1、finalized=1。该终局与视频 14.72 s 所见的持杆站立、洞口不露球一致。"
+     },
+     {
+      "t": 14.7,
+      "log_step": 103,
+      "text_zh": "推断：log step 80 重置后进行的第三回合才是视频中的评分回合；log step 96–97 记录挥杆前后控制步 975–1035，log step 103 保存同一成功动作并明确每控制步 dt=0.01 s，故对应视频 9.75–10.35 s，已用视频 10.04–10.40 s 的击球及白球离开原位核对。log step 102 的终局控制步 1473 对应视频约 14.73 s，与末帧视频 14.72 s 相差一个帧间隔以内；将包含洞内停稳验收的节点达成锚点取为视频约 14.7 s。前两回合的跌倒和推压判罚不属于此评分回合。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 10.04–10.16 s 机器人右手所持推杆的杆头击打白球，随后白球独立滚向球洞。候选方式为“击打”，现有接触方式没有对应项；仅记录杆头对白球的工具链，不重复记录右手握杆。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–14.72 s 机器人由双脚接触果岭支撑身体，击球及等待白球入洞期间保持站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": null,
+ "summary_zh": "评分回合中，机器人右手持杆于视频约 10.1 s 击球，白球滚至洞口后于视频约 13.2 s 起不再露出洞沿；现场裁判在 log step 102 确认成功，唯一节点“一杆推球入洞”于视频约 14.7 s 完成，终态为 1/1。视频对应第三回合，重置前两次失败不计入本次进度或失败标签；成功参考视频也呈现白球在洞口消失的远景外观，仅用于比较。缺少状态日志和洞内近景，毫米级停稳及毫秒级接触无法从视频独立核验，依据现场裁判验收；四项原生指标均未报告数值，保留 null。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "唯一目标节点 ball_holed 于视频约 14.7 s 达成、终态 1/1，与评分器字段（live_success=1）及帧所见一致；控制步映射（dt=0.01 s）有击球事件核验，接触标注合理，未发现需改动的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:55:22Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--jq97mv4/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--jq97mv4/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "fd6bd6eec0eca6c20a9a7148584f59df97c4aad2c6eb8f74d642079eb4b802ce",
+   "video_sha256": "64b3466ad465cf5bf9380ddda4f93beac705d6a379ed8b0873bd2803c2bd902f",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--ngkcwly.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--ngkcwly.json
new file mode 100644
index 0000000000000000000000000000000000000000..c2e7db9f9fbdcb3d57f2053b40cfbedc86869ff0
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--ngkcwly.json
@@ -0,0 +1,214 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-standard--ngkcwly",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 3000,
+   "sim_steps": 19921,
+   "calls": 374,
+   "rows": 348,
+   "resets": 50,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 4630.1,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.825,
+   "peak_processes": 10,
+   "peak_disk_gb": 1.011,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-standard--ngkcwly/replay.mp4",
+  "duration_s": 30.04,
+  "fps": 25.0,
+  "frames": 751,
+  "sha256": "380536d68a95225c4767c939e700cd793bf00fb31dac77537efb478fe9d882ea"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.68,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–1.56 s 逐帧画面中，机器人双脚支撑、右手持推杆转腰挥杆并抬起杆头，白球在脚前原处保持可见，未出现被击出后滚向远处球洞的过程。"
+     },
+     {
+      "t": 30.0,
+      "log_step": null,
+      "text_zh": "观察：视频 1–29 s 的逐秒画面及视频 29.04–30.00 s 的密集末秒画面中，白球仍在机器人脚前，远处浅色条带上的洞口未接收到球；入洞条件全程未成立。"
+     },
+     {
+      "t": null,
+      "log_step": 385,
+      "text_zh": "观察：log step 385 的服务状态为 episode=51、control_steps=3000、done=True、reason=failed、resets_left=0；同一步输出的最终控制器说明记载末次伸肘试杆未击中球。agent 在 log step 384、388 也自述末次回合未记录 stroke、阶段为 0/2；此前会话中的合法击球不计入本评分视频。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "右手包握推杆握柄，持杆完成转腰挥动及随后悬持；左手下垂未辅助握杆。仅记录实际持杆接触，未把未发生的杆头击球记为实际接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "机器人双脚接触果岭、支撑身体完成挥杆和随后的站立，未见倒地。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "ball_holed",
+   "t": 0.68,
+   "log_step": 377,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 0.52–0.80 s，机器人右手带动推杆随转腰从白球附近挥过，随后抬杆；白球没有离开脚前原位，后续直到视频 30.00 s 仍未被送向球洞。log step 377 执行末次重置后的控制器，log step 385 的最终服务状态确认该回合已经失败。"
+  },
+  "unmet": {
+   "ball_holed": false
+  },
+  "phenomena": [
+   {
+    "cat": "接触",
+    "item": "没建立接触",
+    "t": 0.68,
+    "log_step": null,
+    "evidence_zh": "观察：视频 0.00–1.56 s 逐帧可见右手持杆挥动，杆头从白球附近经过后升起，而白球保持在原位；视频 29.04–30.00 s 白球仍留在脚前。推断：本次挥杆未建立所需的杆头与白球击打接触。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "log_step": 385,
+    "evidence_zh": "推断：log step 385 输出的最终脚本用预设转腰、腕部补偿和伸肘命令组合挥杆（ELBOW_PUSH=0.12）；视频约 0.52–0.80 s 实际挥杆没有使杆头接触静止白球。agent 自述伸肘改变了挥杆路径并造成末次挥空（log step 384），与视频相符，支持击球运动控制失败；不足以确认具体控制子类。",
+    "source_zh": "视频 / 日志 / agent 自述"
+   }
+  ],
+  "gap_zh": "没有状态日志，远景无法测出杆头相对球心的三维偏差及瞬时接触力，不能确定挥空主要来自预设轨迹、持杆几何变化还是跟踪误差。run.json 的评分字段未报告四项原生指标，故其 value 均为 null；末次调用的详细服务返回文本也未保留，阶段 0/2 与裁判 time 来自 agent 自述，仅作为视频判定的辅助。"
+ },
+ "summary_zh": "评分采用最后一次现场回合，机器人右手持杆并完成挥杆，但白球始终留在脚前，“一杆推球入洞”全程未达成，最终为 0/1。断点在视频约 0.68 s 的挥空，现象为没建立接触，证据支持控制失败；此前会话中击出白球的调试回合不计入本次进度。缺少状态日志及完整末次服务返回，无法进一步确认挥空的控制子类。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "评分回合为末次现场回合（视频 30.04 s，log step 385 为 failed/3000 步），视频中球始终在脚前、未入洞，ball_holed=false、断点与\"没建立接触\"·控制均有视频和服务读数支持；未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:55:33Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--ngkcwly/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--ngkcwly/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "5e08f2de87cae157fac638c618adc986a3ee93beacb6edbd51b7d01f9b45dd46",
+   "video_sha256": "380536d68a95225c4767c939e700cd793bf00fb31dac77537efb478fe9d882ea",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--nrnbbmc.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--nrnbbmc.json
new file mode 100644
index 0000000000000000000000000000000000000000..26bdd1228b391c4b39123a7e3ebcac35fcd66558
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--nrnbbmc.json
@@ -0,0 +1,220 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-standard--nrnbbmc",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 926,
+   "sim_steps": 38582,
+   "calls": 124,
+   "rows": 38207,
+   "resets": 42,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 7151.3,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.64,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.329,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-standard--nrnbbmc/replay.mp4",
+  "duration_s": 9.28,
+  "fps": 25.0,
+  "frames": 232,
+  "sha256": "1b8d5a7cd6b5fab0d154b4f34b085a155034d0071061a713171d2cec10f7d7e8"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 6.16,
+      "log_step": null,
+      "text_zh": "观察：视频 0.0–5.5 s，机器人右手持推杆抬起、移到白球后方并下放；视频 6.08–6.24 s，杆头挥过脚前白球处，白球离开原位向远处洞口滚动。"
+     },
+     {
+      "t": 9.24,
+      "log_step": null,
+      "text_zh": "观察：视频 6.4–8.2 s，白球滚向浅色条带上的洞口并逐渐减速；视频 8.2–9.24 s，白球停留在条带之前的果岭上，球与洞口明显分离，未落入洞中。"
+     },
+     {
+      "t": null,
+      "log_step": 236,
+      "text_zh": "观察：log step 236 的脚本先重置再提交最后一次 stand 控制序列，并从机器人服务 get_obs 的回合结果元数据读取到 FINAL ATTEMPT OUTCOME: the ball came to rest outside the hole；该结果与视频末态相符。"
+     },
+     {
+      "t": null,
+      "log_step": null,
+      "text_zh": "观察：成功参考视频 inputs/oracle/replay.mp4 的视频 4.0–5.6 s 显示白球到达浅色条带的洞口并消失，参考视频 6.64 s 洞外没有白球；本次视频 9.24 s 则仍能看到洞前白球。参考仅用于辨认入洞的可见表现。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "推断：视频 6.08–6.24 s，机器人右手持杆挥过白球，白球随即离开原位，支持推杆杆头实际击中白球；候选方式为“击打”，现有接触方式没有对应项。握杆不另重复记录；远景不能解析毫秒级接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.0–9.24 s，机器人以双脚在果岭上站立支撑，配合上身和右臂完成移杆与挥杆，末态仍站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "ball_holed",
+   "t": 6.12,
+   "log_step": 236,
+   "where": "操作中",
+   "evidence_zh": "推断：视频约 6.12 s 的这次挥杆击出白球，但视频 8.2–9.24 s 白球在到达洞口前停住，入洞步骤未完成且没有后续修复。log step 236 的最后一次回合由机器人服务明确判为白球停在洞外；锚点取击球时刻，未将日志的 agent 时间换算为视频时间。"
+  },
+  "unmet": {
+   "ball_holed": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 8.2,
+    "log_step": 236,
+    "evidence_zh": "观察：视频 6.16–8.2 s，白球被击出后朝洞口方向滚动，却在浅色洞口条带之前减速停下；视频 8.2–9.24 s 白球一直露在果岭上，与洞口有明显间隔。log step 236 的服务结果亦为白球停在洞外，表现为击球落点偏短。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 6.12,
+    "log_step": 209,
+    "source_zh": "日志",
+    "evidence_zh": "推断：log step 194 恢复挥杆行程越过触球位置 0.12 米及手部高度补偿 0.035 米，log step 209 将脚本的加速度和加速时长设为 9.0 m/s²、0.20 秒；log step 236 用该脚本生成动作并提交最终回合。视频 6.16–9.24 s 显示它建立了击球接触，却未使白球滚到洞口，支持挥杆控制未能产生所需击球效果；这些脚本参数不是实测杆头速度，证据不足以细分控制子类。"
+   }
+  ],
+  "gap_zh": "没有状态日志，远景无法精确测出杆头和白球速度、触球部位、接触时长、右手接触中断及关节跟踪误差；无法确定偏短具体来自速度设定、触球几何还是跟踪误差。agent 在 log step 237 自述的速度饱和及行程限制未获独立测量验证，不据此细分根因。评分器未报告四项原生指标的数值，均记 null。"
+ },
+ "summary_zh": "评分采用 log step 236 的最后一次现场回合：机器人持杆站立、挥杆击出白球，但视频 8.2–9.24 s 白球停在洞前果岭上，一杆入洞节点始终未达成，最终为 0/1。断点在一杆推球入洞的操作中，现象是击球落点偏短，根因保守归为控制。缺少状态及接触、速度和跟踪测量，无法进一步确定具体控制问题。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频 6.1 s 挥杆击球、白球在洞前停下（视频 8.2–9.24 s），ball_holed 未达成，final=false、断点、位姿不准现象与保守的控制根因均与帧和 log step 236 的服务结果相符，未发现需改动之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:47:31Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--nrnbbmc/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--nrnbbmc/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "cb567c97dd52033cfda53cd23064e15fa411a61a90196b9f03832a0eea79a9f5",
+   "video_sha256": "1b8d5a7cd6b5fab0d154b4f34b085a155034d0071061a713171d2cec10f7d7e8",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--qyhcyw7.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--qyhcyw7.json
new file mode 100644
index 0000000000000000000000000000000000000000..590953aff39ce15259aee1d88cc828f417be3dda
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--qyhcyw7.json
@@ -0,0 +1,217 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-standard--qyhcyw7",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 690,
+   "sim_steps": 5482,
+   "calls": 15,
+   "rows": 4282,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 864.3,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.249,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.345,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-standard--qyhcyw7/replay.mp4",
+  "duration_s": 6.92,
+  "fps": 25.0,
+  "frames": 173,
+  "sha256": "f94ee4e324aa849946b8f9042adbbe5c9d1d2aa0536c6c781523a6af7253c131"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.76,
+      "log_step": 171,
+      "text_zh": "观察：视频 2.40–2.68 s 机器人右手持杆，白球仍在脚前；视频约 2.72–2.80 s 杆头经过白球位置后，白球离开原位向远处球洞滚动，机器人双脚持续支撑身体。"
+     },
+     {
+      "t": 5.8,
+      "log_step": 171,
+      "text_zh": "观察：视频 5.00–5.96 s 白球逐渐接近洞口近侧并停下，视频 5.80–6.88 s 的逐帧洞口裁剪仍能看到洞口近侧的亮白球体，没有完整落洞消失。log step 171 是最后现场回合 episode=6 的执行结果，服务返回 stages_reached=1、stages_total=2、outcome=\"the ball came to rest outside the hole\"；该回合 control_steps=690、sim_time_s=6.9，与本视频 6.92 s 时长相符。"
+     },
+     {
+      "t": 6.88,
+      "log_step": 194,
+      "text_zh": "观察：成功参考视频约 5.4–6.6 s 的洞口不再露出白球，本回合视频 6.88 s 洞前仍可见白球。log step 193 后来把默认挥杆时长改为 0.19 s，但 log step 194 运行时服务显示 episode=6 已结束且 resets_left=0，脚本直接退出，没有新的击球回合；该修改不属于本段评分视频。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 2.72–2.80 s，机器人右手持推杆，杆头经过白球后白球向洞口滚出；工具链为推杆杆头作用于白球。候选接触名为“击打”，现有方式中没有对应项，故用其他，不重复记录握杆接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–6.88 s，机器人双脚接触果岭，站立支撑身体完成后摆、击球与随挥。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "ball_holed",
+   "t": 5.8,
+   "log_step": 171,
+   "where": "操作中",
+   "evidence_zh": "观察：机器人在视频约 2.76 s 击出白球，视频约 5.8 s 白球停在洞口近侧，直至视频 6.88 s 仍未落入洞内；log step 171 的现场服务结果明确为球在洞外停稳，仅完成击球阶段。因此失败步骤是一杆推球入洞，锚点取白球在洞前停下的位置。"
+  },
+  "unmet": {
+   "ball_holed": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 5.8,
+    "log_step": 171,
+    "evidence_zh": "观察：视频 2.76–5.8 s 白球沿果岭向洞口滚动后停在洞口近侧，视频 5.80–6.88 s 仍露出白球，没有进入洞内；log step 171 服务判定“the ball came to rest outside the hole”。推断：此次推球距离不足，球的终点未达到入洞位置。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 2.76,
+    "log_step": 167,
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "观察：log step 155、157 中 agent 的计算预测挥杆时长 0.20 s 可以滚过约三米，log step 167 的脚本据此生成固定关节轨迹，log step 171 确实以 T_swing=0.2s 执行；视频 2.76–5.8 s 显示球实际停在洞前，服务确认洞外停球。推断：实际击球控制未给白球足够的前进效果，预先估计的挥杆与滚动关系未达到目标；缺少杆头速度、接触冲量和关节跟踪记录，无法进一步区分力度参数标定与运动跟踪误差，根因保留在控制大项。"
+   }
+  ],
+  "gap_zh": "没有状态日志，现有评分视频仅含场地远景，无法独立核对毫米级洞内高度、停稳位移和毫秒级握杆/杆头接触；洞外停球有视频及最后现场回合的服务判定共同支持。run.json 的 reward 未报告 DAG 所列四项原生指标，均填 null；agent 在 log step 172 的像素反投影是其自算估计，未用作评分器指标。缺少实际击球速度、接触冲量及跟踪误差，控制子原因不能细分；修改为 0.19 s 后没有执行，无法验证修正效果。"
+ },
+ "summary_zh": "机器人右手持杆完成击球，白球滚向洞口后停在洞前，一杆推球入洞节点始终未达成，最终为 0/1。视频约 5.8 s 至结尾仍可见洞前白球，log step 171 的最后现场回合明确判定洞外停球；推断实际挥杆控制的前进效果不足。之后将挥杆时长改为 0.19 s 的脚本没有执行，且缺少状态和接触数据，不能验证修正效果或进一步细分控制原因。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标签与证据一致：视频 5.0 s 起白球停在洞口近侧，log step 171 服务判定 the ball came to rest outside the hole（stages 1/2），ball_holed 未达成、断点与位姿不准/控制归因有证据支持且推断已标注。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:55:03Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--qyhcyw7/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--qyhcyw7/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "d67a00d625526e280172835926a17d0ec8640dce836e746698b98fb5fd31e4c8",
+   "video_sha256": "f94ee4e324aa849946b8f9042adbbe5c9d1d2aa0536c6c781523a6af7253c131",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--vcydxbu.json b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--vcydxbu.json
new file mode 100644
index 0000000000000000000000000000000000000000..f0cba6c45e0e310d393a8e74181cbcc1d4b1f6db
--- /dev/null
+++ b/phase2/embodied-olympics-fun/golf_putt/trajectories/embodied-olympics-fun-golf-putt-i00-standard--vcydxbu.json
@@ -0,0 +1,196 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-golf-putt-i00-standard--vcydxbu",
+ "taskKey": "embodied-olympics-fun/golf_putt",
+ "dag_sha256": "74a1c00f893c3316873e4cc2017a91bda4a2926d26f9ae1a009d7dbb3a3e0854",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "live_success": 1,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 1312,
+   "sim_steps": 8169,
+   "calls": 40,
+   "rows": 2695,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2040.9,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.076,
+   "peak_processes": 11,
+   "peak_disk_gb": 0.631,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-golf-putt-i00-standard--vcydxbu/replay.mp4",
+  "duration_s": 13.16,
+  "fps": 25.0,
+  "frames": 329,
+  "sha256": "f99f890a0cb35f5f532f7961f3086bd8a9f1905eb662844fa86bf199f0acb31c"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "ball_holed": {
+    "method": "visual",
+    "intervals": [
+     [
+      13.12,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.4,
+      "log_step": 82,
+      "text_zh": "观察：视频 9.28–9.64 s 中，机器人双脚站在果岭上，右手持推杆向前挥动；杆头经过脚前白球位置后，白球离开原位并向远处浅色条带上的洞口滚动。"
+     },
+     {
+      "t": 11.2,
+      "log_step": null,
+      "text_zh": "观察：视频 10.80–11.52 s 的洞口区域逐帧检查可见白球接近并进入洞口区域，随后不再能辨出洞外白球；视频 11.60–13.12 s 未见白球重新露出，机器人仍站立持杆。"
+     },
+     {
+      "t": 13.12,
+      "log_step": 82,
+      "text_zh": "观察：log step 82 在 reset 后明确进入 episode 6，最终打印的机器人服务返回值为 done=True、reason=success、outcome=success、stages_reached=2，回合仿真时间约 13.12 s；log step 83 的服务 status 再次报告 episode=6、control_steps=1312、done=True、reason=success。评分字段 live_success=1、graded_by_live=1，说明评分采用这次最终现场回合，前五次重置回合不属于本视频。"
+     },
+     {
+      "t": 13.12,
+      "log_step": 83,
+      "text_zh": "推断：log step 3 的服务规格给出控制周期 0.01 s，log step 83 的最终回合控制步 1312 对应视频约 13.12 s；视频总长 13.16 s，且脚本在回合仿真 9.0 s 开始挥杆、约 9.33 s 经过摆动底点，与视频 9.32–9.44 s 的击球相符，故采用视频 13.12 s 作为裁判完成入洞停稳验收的节点起点。远景只支持球进洞后不再露出的判断，全球低于果岭及洞内停稳 1.0 秒由本次服务成功结果确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 9.32–9.64 s，机器人右手持推杆，杆头经过白球后白球立即向洞口滚动；候选接触名为“击打”，现有接触方式没有对应项。此处记录杆头对白球的工具链，不重复记录握杆。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.0–13.12 s，机器人双脚在果岭上支撑身体，击球及等待白球滚入洞时均保持站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "ball_to_hole_m",
+   "unit": "m",
+   "threshold": 0.054,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "ball_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "head_contact_s",
+   "unit": "s",
+   "threshold": 0.03,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": null,
+ "summary_zh": "本次评分视频对应最终第六回合：机器人右手持推杆，在视频约 9.4 s 击球，白球滚至洞口后不再露出；服务在视频约 13.12 s 对应的回合终点确认成功，唯一节点一杆推球入洞达成，最终为 1/1。前五次重置回合的失败不属于本次评分轨迹。没有状态日志，远景不能独立量测洞内高度、毫米级停稳或毫秒级接触，精细验收依据本次裁判成功结果；评分字段未报告四项原生指标数值，均保留为 null。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "单节点 ball_holed 达成、final=true 与评分成功一致；击球时间约 9.4 s、球入洞后不再露出（约 11 s 起）与帧相符，13.12 s 的节点起点由服务控制步 1312 的推断得出且标注为推断，contact_actual 合理。未发现需改动处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:48:50Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--vcydxbu/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-golf-putt-i00-standard--vcydxbu/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "f3bd5e263b6f5863b9e1eb6c098d5df7f54ca74f1a308536bc543ab2013d87da",
+   "video_sha256": "f99f890a0cb35f5f532f7961f3086bd8a9f1905eb662844fa86bf199f0acb31c",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--8gsjzxu.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--8gsjzxu.json
new file mode 100644
index 0000000000000000000000000000000000000000..8dd018bfaa358a51165ba375d370c28752281e7b
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--8gsjzxu.json
@@ -0,0 +1,313 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-privileged--8gsjzxu",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 514,
+   "sim_steps": 514,
+   "horizon": 6000,
+   "episode_over": 1,
+   "progress": 0.0,
+   "stages_reached": 0,
+   "stages_total": 10,
+   "fell": 0,
+   "replay_s": 4.43,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 0.099,
+   "metric_bounces": 0.0,
+   "metric_touches": 0.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 476,
+   "agent_resets": 577,
+   "agent_replays": 7,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 9.16,
+   "peak_processes": 24,
+   "peak_disk_gb": 0.24,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-privileged--8gsjzxu/replay.mp4",
+  "duration_s": 5.16,
+  "fps": 25.0,
+  "frames": 129,
+  "sha256": "300dfccd482e55bd29b05ac0279cbe07864c04c2754f249c8b3184f3ecbd3f5f"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.84,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–4.80 s 机器人将足球保持在双掌之间，约视频 4.20 s 起抬右脚，视频 4.84–4.96 s 双掌分开、足球下落；视频 5.00–5.12 s 球从抬起的右脚前方落向草地，未出现脚击球后离脚上升，机器人仍站立。"
+     },
+     {
+      "t": 5.12,
+      "log_step": 333,
+      "text_zh": "观察：评分器字段 stages_reached=0、metric_touches=0、metric_bounces=0、fell=0；log step 239、277、333 对交付的同一514步文件回放均返回 touches=0、stages=[]、足球触地结束，末态哈希一致，与视频 5.12 s 的落球相符。第1次有效颠球未完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "视频 0.00–4.80 s 左右平掌共同夹持足球，随后分开让球落下；候选名称“双掌挤压”。两只平掌均未包围足球，现有方式无合适的双掌挤压项，不能记作抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "准备首次颠球时双脚承重接触草地；视频约 4.20–5.12 s 右脚抬起，左脚继续在草地支撑身体。右脚没有击中下落足球，因此不记录脚对球的击打接触。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.12,
+      "log_step": 333,
+      "text_zh": "推断：视频 0.00–5.12 s 只有双掌持球、抬脚和一次放球下落，至视频 5.12 s 未出现任何脚击后上升；评分器 stages_reached=0、metric_touches=0，log step 333 回放交付文件也为 touches=0、stages=[]。第1次尚未完成，第2次有效颠球未开始且未达到。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.12,
+      "log_step": 333,
+      "text_zh": "推断：视频 0.00–5.12 s 只有双掌持球、抬脚和一次放球下落，至视频 5.12 s 未出现任何脚击后上升；评分器 stages_reached=0、metric_touches=0，log step 333 回放交付文件也为 touches=0、stages=[]。第1次尚未完成，第3次有效颠球未开始且未达到。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.12,
+      "log_step": 333,
+      "text_zh": "推断：视频 0.00–5.12 s 只有双掌持球、抬脚和一次放球下落，至视频 5.12 s 未出现任何脚击后上升；评分器 stages_reached=0、metric_touches=0，log step 333 回放交付文件也为 touches=0、stages=[]。第1次尚未完成，第4次有效颠球未开始且未达到。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.12,
+      "log_step": 333,
+      "text_zh": "推断：视频 0.00–5.12 s 只有双掌持球、抬脚和一次放球下落，至视频 5.12 s 未出现任何脚击后上升；评分器 stages_reached=0、metric_touches=0，log step 333 回放交付文件也为 touches=0、stages=[]。第1次尚未完成，第5次有效颠球未开始且未达到。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.12,
+      "log_step": 333,
+      "text_zh": "推断：视频 0.00–5.12 s 只有双掌持球、抬脚和一次放球下落，至视频 5.12 s 未出现任何脚击后上升；评分器 stages_reached=0、metric_touches=0，log step 333 回放交付文件也为 touches=0、stages=[]。第1次尚未完成，第6次有效颠球未开始且未达到。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.12,
+      "log_step": 333,
+      "text_zh": "推断：视频 0.00–5.12 s 只有双掌持球、抬脚和一次放球下落，至视频 5.12 s 未出现任何脚击后上升；评分器 stages_reached=0、metric_touches=0，log step 333 回放交付文件也为 touches=0、stages=[]。第1次尚未完成，第7次有效颠球未开始且未达到。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.12,
+      "log_step": 333,
+      "text_zh": "推断：视频 0.00–5.12 s 只有双掌持球、抬脚和一次放球下落，至视频 5.12 s 未出现任何脚击后上升；评分器 stages_reached=0、metric_touches=0，log step 333 回放交付文件也为 touches=0、stages=[]。第1次尚未完成，第8次有效颠球未开始且未达到。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.12,
+      "log_step": 333,
+      "text_zh": "推断：视频 0.00–5.12 s 只有双掌持球、抬脚和一次放球下落，至视频 5.12 s 未出现任何脚击后上升；评分器 stages_reached=0、metric_touches=0，log step 333 回放交付文件也为 touches=0、stages=[]。第1次尚未完成，第9次有效颠球未开始且未达到。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.12,
+      "log_step": 333,
+      "text_zh": "推断：视频 0.00–5.12 s 只有双掌持球、抬脚和一次放球下落，至视频 5.12 s 未出现任何脚击后上升；评分器 stages_reached=0、metric_touches=0，log step 333 回放交付文件也为 touches=0、stages=[]。第1次尚未完成，第10次有效颠球未开始且未达到。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "value": 0.0,
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better"
+  },
+  {
+   "name": "ball_height_m",
+   "value": 0.099,
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_1",
+   "t": 5.0,
+   "log_step": 333,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 4.84–5.12 s 机器人已放开双掌并抬起右脚，但足球在右脚前方持续下落，视频 5.00 s 附近已错过脚部截击，未再升起；视频 5.12 s 球到达草地，机器人未倒下。log step 333 的交付文件回放返回足球触地结束、touches=0，故失败发生于第1次颠球操作。"
+  },
+  "unmet": {
+   "juggles_counted_1": false,
+   "juggles_counted_2": false,
+   "juggles_counted_3": false,
+   "juggles_counted_4": false,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 5.0,
+    "log_step": null,
+    "evidence_zh": "观察：视频 4.84–5.12 s 足球脱离双掌后向下运动，机器人右脚保持抬起但没有跟到球的下落位置；视频 5.00–5.12 s 球从右脚前方落到草地，整段没有向上的脚击反弹。机器人未能截击这个运动中的目标。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "轨迹或逆解",
+    "source_zh": "日志",
+    "evidence_zh": "推断：agent 的控制脚本未生成可执行并能把右脚送到落球处的关节轨迹。log step 245 对默认控制器的诊断复现514步触地失败，逆解欲将脚部送至[0.39,0.06,0.32]米，但实际右脚锚点约为[0.015,0.056,0.265]米；log step 248 的会话诊断在仿真 4.60–4.90 s 给出右踝滚转目标约−0.645至−0.692弧度，实际仅约−0.263至−0.269弧度，log step 250 读取模型右踝滚转限位为[−0.262,0.262]弧度，目标超出范围。log step 238 生成交付文件时在仿真 5.01 s 输出球心x=0.349米、右脚标记x=0.014米；视频 4.84–5.12 s 同样显示脚留在落球后方。上述诊断支持逆解及轨迹可执行性问题，不采用 agent 关于所有合法方案均不可行的自述。"
+   }
+  ],
+  "gap_zh": "没有状态日志；25帧每秒的固定视角不能独立解析毫秒级接触、精确速度和关节受力。节点未达由完整落球过程及交付回放的原生计数共同确定；根因的限位和关节跟踪依据是 log step 245、248、250 的会话诊断，不能据此确定交付回放每一帧的电机饱和或量化各控制环节的贡献。"
+ },
+ "summary_zh": "机器人完成了首次颠球的抬脚和放球准备，但足球在视频 4.84–5.12 s 从右脚前方直接落地，十个有效颠球节点均未达到，评分器计数为0且未跌倒。断点为第1次颠球，现象是未能截击下落足球；日志诊断支持逆解目标超出右踝限位、脚位未到落球处的控制问题。没有状态日志，无法独立核验毫秒接触及精确受力，会话中其他测试的跌倒和碰腿不计入本次交付。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频 0–5.12 s 球一直在双掌间，随后落地，无脚击上升；评分器 stages_reached=0、touches=0，log step 333 回放证实。十个节点均未达、断点 juggles_counted_1、现象与根因（逆解/轨迹，右踝滚转目标超限位）均有 log step 245/248/250 的数值支持，未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:48:33Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--8gsjzxu/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--8gsjzxu/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "f46d6bee95747aead1012268df279698384da2271af232f6afc44507c551524f",
+   "video_sha256": "300dfccd482e55bd29b05ac0279cbe07864c04c2754f249c8b3184f3ecbd3f5f",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--kihvqab.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--kihvqab.json
new file mode 100644
index 0000000000000000000000000000000000000000..4320b0bb81f04659e956cb9151b1576331c57943
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--kihvqab.json
@@ -0,0 +1,324 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-privileged--kihvqab",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.1,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 85,
+   "sim_steps": 85,
+   "horizon": 6000,
+   "episode_over": 0,
+   "progress": 0.1,
+   "stages_reached": 1,
+   "stages_total": 10,
+   "fell": 0,
+   "replay_s": 6.01,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 0.851,
+   "metric_bounces": 0.0,
+   "metric_touches": 1.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 622,
+   "agent_resets": 16,
+   "agent_replays": 5,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.744,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.277,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-privileged--kihvqab/replay.mp4",
+  "duration_s": 0.88,
+  "fps": 25.0,
+  "frames": 22,
+  "sha256": "46ee9802553ad40faa13e345ecc160403f8a0c677f46ca2d67fc9e673110e5c0"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      0.8,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.72,
+      "log_step": null,
+      "text_zh": "观察：已检查视频 0.00–0.84 s 全部22帧；机器人先以双掌夹持足球并张臂放球，球在视频 0.64–0.72 s 下落至右脚附近，在视频 0.76–0.84 s 离脚上升，未见球触地或机器人倒地。"
+     },
+     {
+      "t": 0.8,
+      "log_step": 439,
+      "text_zh": "推断：视频 0.80 s 球已明显离脚升高，取此帧为首次有效颠球的近似达成点（约有一帧误差）；log step 439 对所交付85行轨迹的回放返回 stages=[touch 1]、touches=1、ball_height_m=0.851，与最终评分器 stages_reached=1、metric_touches=1 一致。有效计数由评分器补证，未从25 fps画面独立测算毫秒接触或速度门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "观察：视频 0.00–0.40 s 初始双掌从两侧夹住足球，随后张开释放；候选方式“对向挤压夹持”，双掌未各自包围足球，不属于抓握。此为首次颠球的放球准备。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 0.72–0.80 s 右脚在足球下方摆动，足球随即离脚上升；候选方式“击打”，现有接触方式无击打项。log step 358 同参数会话测试的裁判事件也标记 right_foot，仅作为作用脚身份的辅助。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–0.72 s 双脚先接触草地，右脚摆动击球时左脚承担地面支撑。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": 439,
+      "text_zh": "观察：视频 0.00–0.84 s 全部帧只显示一次下落和踢起；末帧足球仍处于首次击球后的上升段，没有第2次离脚上升。log step 439 对所交付轨迹的回放仅有 touch 1，最终评分器 stages_reached=1、metric_touches=1，故第2次有效颠球未达成。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": 439,
+      "text_zh": "观察：视频 0.00–0.84 s 全部帧只显示一次下落和踢起；末帧足球仍处于首次击球后的上升段，没有第3次离脚上升。log step 439 对所交付轨迹的回放仅有 touch 1，最终评分器 stages_reached=1、metric_touches=1，故第3次有效颠球未达成。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": 439,
+      "text_zh": "观察：视频 0.00–0.84 s 全部帧只显示一次下落和踢起；末帧足球仍处于首次击球后的上升段，没有第4次离脚上升。log step 439 对所交付轨迹的回放仅有 touch 1，最终评分器 stages_reached=1、metric_touches=1，故第4次有效颠球未达成。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": 439,
+      "text_zh": "观察：视频 0.00–0.84 s 全部帧只显示一次下落和踢起；末帧足球仍处于首次击球后的上升段，没有第5次离脚上升。log step 439 对所交付轨迹的回放仅有 touch 1，最终评分器 stages_reached=1、metric_touches=1，故第5次有效颠球未达成。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": 439,
+      "text_zh": "观察：视频 0.00–0.84 s 全部帧只显示一次下落和踢起；末帧足球仍处于首次击球后的上升段，没有第6次离脚上升。log step 439 对所交付轨迹的回放仅有 touch 1，最终评分器 stages_reached=1、metric_touches=1，故第6次有效颠球未达成。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": 439,
+      "text_zh": "观察：视频 0.00–0.84 s 全部帧只显示一次下落和踢起；末帧足球仍处于首次击球后的上升段，没有第7次离脚上升。log step 439 对所交付轨迹的回放仅有 touch 1，最终评分器 stages_reached=1、metric_touches=1，故第7次有效颠球未达成。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": 439,
+      "text_zh": "观察：视频 0.00–0.84 s 全部帧只显示一次下落和踢起；末帧足球仍处于首次击球后的上升段，没有第8次离脚上升。log step 439 对所交付轨迹的回放仅有 touch 1，最终评分器 stages_reached=1、metric_touches=1，故第8次有效颠球未达成。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": 439,
+      "text_zh": "观察：视频 0.00–0.84 s 全部帧只显示一次下落和踢起；末帧足球仍处于首次击球后的上升段，没有第9次离脚上升。log step 439 对所交付轨迹的回放仅有 touch 1，最终评分器 stages_reached=1、metric_touches=1，故第9次有效颠球未达成。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": 439,
+      "text_zh": "观察：视频 0.00–0.84 s 全部帧只显示一次下落和踢起；末帧足球仍处于首次击球后的上升段，没有第10次离脚上升。log step 439 对所交付轨迹的回放仅有 touch 1，最终评分器 stages_reached=1、metric_touches=1，故第10次有效颠球未达成。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better",
+   "value": 1.0
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": 0.851
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_2",
+   "t": null,
+   "log_step": 437,
+   "where": "首步前",
+   "evidence_zh": "观察：视频 0.76–0.84 s 仅出现首次击球后的上升，未出现第二次迎球击打。log step 437 的生成脚本只执行一次放球及右脚击球，在控制步85前结束并保存；log step 674 再运行该生成脚本，log step 680 最终报告仍为85个控制步、Touch 1。最终评分器 episode_over=0、fell=0、stages_reached=1，表明交付在第二次颠球启动前结束，不能把会话测试中的后续跌倒当作本次判负事件。"
+  },
+  "unmet": {
+   "juggles_counted_1": true,
+   "juggles_counted_2": false,
+   "juggles_counted_3": false,
+   "juggles_counted_4": false,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  },
+  "phenomena": [
+   {
+    "cat": "没尝试",
+    "item": "没尝试",
+    "t": 0.84,
+    "log_step": 437,
+    "evidence_zh": "观察：视频 0.84 s 末帧中足球仍从首次击球上升；全段未见第二次击球准备或新的触球周期。log step 437 脚本的最后一段只在首次击球姿态与上挑姿态之间切换，到控制步85即保存，没有第二次颠球段；log step 439 的交付回放仅计入 touch 1。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "log_step": 414,
+    "source_zh": "日志",
+    "evidence_zh": "推断：机器人首次击球后的恢复与持续支撑控制未解决，限制了连续颠球。依据是会话测试 log step 358 首次计入 touch 1 后骨盆持续下降并由服务返回 fell；log step 414 的平滑回收测试仍在控制步100附近跌倒；log step 488 改用张臂双足恢复仍跌倒；更晚的 log step 648 快速回双足支撑测试也只达 touch 1 后跌倒。最终 log step 437/674 交付仍停在首次击球后的85行。以上均为会话证据，不能据此声称视频内发生了跌倒；不足以确定具体控制子类。"
+   }
+  ],
+  "gap_zh": "没有仿真状态日志，25 fps视频不能独立确认接触簇时长、能量速度门槛及精确计数时刻，首次有效计数由最终评分器补证。交付止于首次击球后，未展示足球回落和第二次迎球；控制根因由多次会话续接测试推断，日志未明确说明最终为何仅保留85行，无法进一步确定具体控制子类。"
+ },
+ "summary_zh": "机器人放球后用右脚完成第1次有效颠球，视频约0.80 s 达成，最终为1/10；第2次颠球尚未启动，交付便结束，评分器未记录跌倒或回合终止。会话中多次首次击球后的恢复测试失稳，支持持续控制能力不足的推断，但这些跌倒不属于交付录像；无状态日志，精确触球验收依赖评分器计数。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "第1次颠球约视频0.8 s达成、其余未达成、终分1/10均与视频帧和评分器字段一致；断点在 juggles_counted_2（首步前/没尝试，交付仅85行且止于首次击球）与交付内容相符，会话中的跌倒仅作为控制根因的推断且已声明不属于交付录像，未发现需修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:52:03Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--kihvqab/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--kihvqab/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "4e6722b2782bb12fa0c4ed0bf2c17f81dfe67c028c7cf9ee9d272fe5f715c6ea",
+   "video_sha256": "46ee9802553ad40faa13e345ecc160403f8a0c677f46ca2d67fc9e673110e5c0",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--o6ebpor.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--o6ebpor.json
new file mode 100644
index 0000000000000000000000000000000000000000..b052b7e63b8a0fb3dfde54a3470d473b42e9dc96
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--o6ebpor.json
@@ -0,0 +1,361 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-privileged--o6ebpor",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.3,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 337,
+   "sim_steps": 337,
+   "horizon": 6000,
+   "episode_over": 1,
+   "progress": 0.3,
+   "stages_reached": 3,
+   "stages_total": 10,
+   "fell": 0,
+   "replay_s": 3.79,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 0.088,
+   "metric_bounces": 0.0,
+   "metric_touches": 3.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 1372,
+   "agent_resets": 0,
+   "agent_replays": 11,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 11.217,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.844,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-privileged--o6ebpor/replay.mp4",
+  "duration_s": 3.4,
+  "fps": 25.0,
+  "frames": 85,
+  "sha256": "b0fbdace00fc6d2a208d72a7953cfe65d1507ac12da53c1f8972e86118af1e6a"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 3,
+  "final": 3,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      0.64,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.64,
+      "log_step": 499,
+      "text_zh": "观察：视频 0.24–0.56 s，机器人张开双掌放球，足球下落至左脚；视频 0.56–0.72 s，球从左脚附近转为上升并离脚，视频 0.84 s 已升至腰部附近。log step 499 对交付的337行轨迹回放，裁判列出 touch 1（left_foot）；评分器最终 stages_reached=3。首次达到时间按视频取约0.64 s，毫秒接触和速度合规由该回放裁判计数补充。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "观察：视频 0.00–0.24 s，机器人双掌相对挤压足球，随后分开双掌放落足球；候选名“双掌挤压”，平掌没有合围抓握，现有方式无准确对应项。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：机器人左脚短促击打足球后球离脚上升；候选接触名为“击打”，现有接触方式无对应项，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：左脚击球时，机器人右脚接触草地承重支撑身体。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      1.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 1.24,
+      "log_step": 499,
+      "text_zh": "观察：视频 1.04–1.20 s，足球再次下落，机器人抬起右脚承接；视频 1.24–1.52 s，球离开右脚持续向上，升至头部附近。log step 499 对交付文件的回放裁判列出 touch 2（right_foot）；按画面取视频约1.24 s达到，已计入的第1次保留。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：机器人右脚短促击打足球后球离脚上升；候选接触名为“击打”，现有接触方式无对应项，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：右脚击球时，机器人左脚接触草地承重支撑身体。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.2,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.2,
+      "log_step": 499,
+      "text_zh": "观察：视频 1.76–2.12 s，足球从高处下落到右脚前，机器人屈伸右膝击球；视频 2.16–2.40 s，球从右脚附近快速离脚上升并超过腰部。log step 499 对交付文件的回放裁判列出 touch 3（right_foot），最终 touches=3；按画面取视频约2.20 s达到。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：机器人右脚短促击打足球后球离脚上升；候选接触名为“击打”，现有接触方式无对应项，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：右脚击球时，机器人左脚接触草地承重支撑身体。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.2,
+      "log_step": 499,
+      "text_zh": "观察：视频 2.40–3.36 s，机器人放低击球腿并调整站姿，足球在其侧方下落；视频 3.20–3.36 s，双脚均未移到下落足球下方，球没有第四次离脚上升，最后接近并落到草地。log step 499 的交付回放以“the ball touched the ground”结束，只有 touch 1–3，最终评分器 metric_touches=3。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.40–3.36 s，机器人双脚落地并调整支撑站姿以接续颠球；下落的足球位于脚侧方，没有第四次脚球接触。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.36,
+      "log_step": 499,
+      "text_zh": "观察：视频 3.20–3.36 s，足球在机器人脚侧落地，录像结束前没有后续击起。推断：结合交付回放（log step 499）仅列出 touch 1–3，以及最终评分器 stages_reached=3、metric_touches=3，顺序计数的第5次颠球未完成，且未进入该次击球。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.36,
+      "log_step": 499,
+      "text_zh": "观察：视频 3.20–3.36 s，足球在机器人脚侧落地，录像结束前没有后续击起。推断：结合交付回放（log step 499）仅列出 touch 1–3，以及最终评分器 stages_reached=3、metric_touches=3，顺序计数的第6次颠球未完成，且未进入该次击球。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.36,
+      "log_step": 499,
+      "text_zh": "观察：视频 3.20–3.36 s，足球在机器人脚侧落地，录像结束前没有后续击起。推断：结合交付回放（log step 499）仅列出 touch 1–3，以及最终评分器 stages_reached=3、metric_touches=3，顺序计数的第7次颠球未完成，且未进入该次击球。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.36,
+      "log_step": 499,
+      "text_zh": "观察：视频 3.20–3.36 s，足球在机器人脚侧落地，录像结束前没有后续击起。推断：结合交付回放（log step 499）仅列出 touch 1–3，以及最终评分器 stages_reached=3、metric_touches=3，顺序计数的第8次颠球未完成，且未进入该次击球。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.36,
+      "log_step": 499,
+      "text_zh": "观察：视频 3.20–3.36 s，足球在机器人脚侧落地，录像结束前没有后续击起。推断：结合交付回放（log step 499）仅列出 touch 1–3，以及最终评分器 stages_reached=3、metric_touches=3，顺序计数的第9次颠球未完成，且未进入该次击球。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.36,
+      "log_step": 499,
+      "text_zh": "观察：视频 3.20–3.36 s，足球在机器人脚侧落地，录像结束前没有后续击起。推断：结合交付回放（log step 499）仅列出 touch 1–3，以及最终评分器 stages_reached=3、metric_touches=3，顺序计数的第10次颠球未完成，且未进入该次击球。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "value": 3.0,
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better"
+  },
+  {
+   "name": "ball_height_m",
+   "value": 0.088,
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_4",
+   "t": 3.2,
+   "log_step": 499,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 2.40–3.36 s，机器人在第三次击球后收腿、调整站姿，但没跟上足球的下落路径；视频 3.20 s 球已在双脚侧方下落，视频 3.36 s 球落到草地，机器人仍站立。交付文件回放（log step 499、501）均只有前三次计数并以球触地结束，最终评分器 stages_reached=3、fell=0。推断：未修复的断点为接续第4次颠球。"
+  },
+  "unmet": {
+   "juggles_counted_1": true,
+   "juggles_counted_2": true,
+   "juggles_counted_3": true,
+   "juggles_counted_4": false,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 3.2,
+    "log_step": 499,
+    "evidence_zh": "观察：视频 2.96–3.36 s，足球从高处落向机器人侧方，机器人调整双脚但没有跟随到球下方，球直接落地，未出现第四次脚部向上击球。推断：机器人错过了运动中足球的接续击球窗口；此为同一次未跟上的失误。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 3.2,
+    "log_step": 477,
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "观察：log step 477 的脚本替换内容显示，保存生成器在第三次计数后持续使用零平移、零转向目标的 walk_falcon 恢复命令直至回合结束；该步将追球命令写入另一个会话测试文件，未替换保存生成器。log step 499、501 对交付文件的回放均为相同337行、相同末态哈希、三次计数后球触地。视频 2.40–3.36 s 显示机器人恢复站姿但球与脚的水平间隔增大，始终未在下落时补上击球。推断：交付控制序列只实现了前三次击球及站姿恢复，未实现随球移动和接续触球；据此归为控制，证据不足以进一步确定控制子类。"
+   }
+  ],
+  "gap_zh": "没有状态日志；25帧每秒录像无法独立核验每次约0.01秒的触球历时及速度门槛，前三次有效性以交付文件回放的裁判事件和最终评分器计数补充。节点达到时刻按已检查的视频帧取近似值，未将agent工作时间或会话测试的时间轴换成视频时间；无法进一步区分接续控制的具体子类。"
+ },
+ "summary_zh": "机器人放球后先用左脚、再连续两次用右脚将球击起，完成第1–3次有效颠球，最终进度为3/10。第4次接续时，机器人恢复站姿却未跟上足球的下落路径，球在视频约3.36 s落地，机器人未跌倒；交付回放确认三次计数后以球触地结束，根因归为控制。没有状态日志，毫秒接触及速度合法性由同一交付文件的裁判回放补充，达到时刻为视频近似值。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点时间与交付回放裁判事件（touch 1/2/3 于 0.634/1.24/2.156 s）及视频帧一致；断点在第4次颠球、未跌倒，现象与根因（控制）有日志和画面支撑，未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:55:43Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--o6ebpor/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--o6ebpor/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "3573e7f826423a57140d32ad42482364c6157254f1f81f3143b1d1042ea26273",
+   "video_sha256": "b0fbdace00fc6d2a208d72a7953cfe65d1507ac12da53c1f8972e86118af1e6a",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--sfy2pjr.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--sfy2pjr.json
new file mode 100644
index 0000000000000000000000000000000000000000..04eada62acaa7c21b14cbba126664e677b1753c4
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--sfy2pjr.json
@@ -0,0 +1,403 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-privileged--sfy2pjr",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.4,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 1497,
+   "sim_steps": 1497,
+   "horizon": 6000,
+   "episode_over": 1,
+   "progress": 0.4,
+   "stages_reached": 4,
+   "stages_total": 10,
+   "fell": 0,
+   "replay_s": 9.09,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 0.354,
+   "metric_bounces": 1.0,
+   "metric_touches": 4.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 146,
+   "agent_resets": 0,
+   "agent_replays": 1,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 2.802,
+   "peak_processes": 21,
+   "peak_disk_gb": 0.253,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-privileged--sfy2pjr/replay.mp4",
+  "duration_s": 15.0,
+  "fps": 25.0,
+  "frames": 375,
+  "sha256": "88cafa21ec8b0de751d89ec8a246406c7104b658eb9227489849f3593e37ef7e"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 4,
+  "final": 4,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      12.52,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 12.52,
+      "log_step": null,
+      "text_zh": "观察：视频 12.36–12.40 s 足球在右脚前低处改变运动方向，视频 12.44–12.64 s 离脚上升。机器人始终以左脚支撑，没有跌倒。"
+     },
+     {
+      "t": 12.52,
+      "log_step": 140,
+      "text_zh": "推断：log step 140 对交付文件的重建验证依次输出 strike 0 touches 0、strike 1 touches 1，直至 strike 4 touches 4；log step 141 的独立验证回放也只列出 touch 1 至 touch 4，且 reward.metric_touches=4、deterministic=1。结合视频 11.84–12.16 s 的先行低反弹和后续四次上升，将本次对应第1次有效颠球；视频 12.52 s 是上升达标的近似锚点，误差约0.1–0.2秒，毫秒接触及速度判据采用裁判计数。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "观察：视频 0–11.20 s 双掌从两侧挤持足球，视频 11.28–11.60 s 双臂前摆并分开释放足球，作为首次颠球的准备；候选接触名为“双掌挤压”，单掌没有包围球，不能标抓握，现有方式无对应项（D12）。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：机器人以右脚短促击打足球后球离脚上升；候选接触名为“击打”，现有接触方式无对应项。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：左脚接触草地承重，在右脚迎球和击球时支撑机器人。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      13.04,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 13.04,
+      "log_step": null,
+      "text_zh": "观察：视频 12.88–12.92 s 足球再次到达右脚处，视频 12.96–13.20 s 离脚上升。机器人始终以左脚支撑，没有跌倒。"
+     },
+     {
+      "t": 13.04,
+      "log_step": 140,
+      "text_zh": "推断：log step 140 对交付文件的重建验证依次输出 strike 0 touches 0、strike 1 touches 1，直至 strike 4 touches 4；log step 141 的独立验证回放也只列出 touch 1 至 touch 4，且 reward.metric_touches=4、deterministic=1。结合视频 11.84–12.16 s 的先行低反弹和后续四次上升，将本次对应第2次有效颠球；视频 13.04 s 是上升达标的近似锚点，误差约0.1–0.2秒，毫秒接触及速度判据采用裁判计数。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：机器人以右脚短促击打足球后球离脚上升；候选接触名为“击打”，现有接触方式无对应项。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：左脚接触草地承重，在右脚迎球和击球时支撑机器人。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [
+     [
+      13.6,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 13.6,
+      "log_step": null,
+      "text_zh": "观察：视频 13.48–13.52 s 足球到达右脚处，视频 13.56–13.84 s 离脚上升至更高处。机器人始终以左脚支撑，没有跌倒。"
+     },
+     {
+      "t": 13.6,
+      "log_step": 140,
+      "text_zh": "推断：log step 140 对交付文件的重建验证依次输出 strike 0 touches 0、strike 1 touches 1，直至 strike 4 touches 4；log step 141 的独立验证回放也只列出 touch 1 至 touch 4，且 reward.metric_touches=4、deterministic=1。结合视频 11.84–12.16 s 的先行低反弹和后续四次上升，将本次对应第3次有效颠球；视频 13.60 s 是上升达标的近似锚点，误差约0.1–0.2秒，毫秒接触及速度判据采用裁判计数。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：机器人以右脚短促击打足球后球离脚上升；候选接触名为“击打”，现有接触方式无对应项。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：左脚接触草地承重，在右脚迎球和击球时支撑机器人。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [
+     [
+      14.32,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 14.32,
+      "log_step": null,
+      "text_zh": "观察：视频 14.20–14.24 s 足球到达右脚处，视频 14.28–14.60 s 明显离脚上升，随后再下落。机器人始终以左脚支撑，没有跌倒。"
+     },
+     {
+      "t": 14.32,
+      "log_step": 140,
+      "text_zh": "推断：log step 140 对交付文件的重建验证依次输出 strike 0 touches 0、strike 1 touches 1，直至 strike 4 touches 4；log step 141 的独立验证回放也只列出 touch 1 至 touch 4，且 reward.metric_touches=4、deterministic=1。结合视频 11.84–12.16 s 的先行低反弹和后续四次上升，将本次对应第4次有效颠球；视频 14.32 s 是上升达标的近似锚点，误差约0.1–0.2秒，毫秒接触及速度判据采用裁判计数。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：机器人以右脚短促击打足球后球离脚上升；候选接触名为“击打”，现有接触方式无对应项。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：左脚接触草地承重，在右脚迎球和击球时支撑机器人。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 14.96,
+      "log_step": 141,
+      "text_zh": "观察：视频 14.72–14.96 s 足球从高处下落至机器人右小腿附近，录像随即结束，没有后续合法离脚上升；log step 141 针对交付文件的验证回放报告 failed=\"the ball touched a part other than a foot\"、stages 仅 touch 1 至 touch 4，reward.stages_reached=4、metric_touches=4。推断：累计第5次有效颠球未达到。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 14.96 s 足球到达右小腿前侧，交付文件验证回放在 log step 141 确认发生非脚部触球；agent 在 log step 142–143 称其为小腿犯规。推断：实际接触为右小腿与足球碰撞，候选接触名“碰撞”，现有方式无对应项；未形成第五次合法脚部击打。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 14.72–14.96 s 左脚继续接触草地承重，右腿准备迎接下落的足球，机器人保持站立。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 14.96,
+      "log_step": 141,
+      "text_zh": "观察：视频 14.72–14.96 s 足球从高处下落至机器人右小腿附近，录像随即结束，没有后续合法离脚上升；log step 141 针对交付文件的验证回放报告 failed=\"the ball touched a part other than a foot\"、stages 仅 touch 1 至 touch 4，reward.stages_reached=4、metric_touches=4。推断：累计第6次有效颠球未达到。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 14.96,
+      "log_step": 141,
+      "text_zh": "观察：视频 14.72–14.96 s 足球从高处下落至机器人右小腿附近，录像随即结束，没有后续合法离脚上升；log step 141 针对交付文件的验证回放报告 failed=\"the ball touched a part other than a foot\"、stages 仅 touch 1 至 touch 4，reward.stages_reached=4、metric_touches=4。推断：累计第7次有效颠球未达到。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 14.96,
+      "log_step": 141,
+      "text_zh": "观察：视频 14.72–14.96 s 足球从高处下落至机器人右小腿附近，录像随即结束，没有后续合法离脚上升；log step 141 针对交付文件的验证回放报告 failed=\"the ball touched a part other than a foot\"、stages 仅 touch 1 至 touch 4，reward.stages_reached=4、metric_touches=4。推断：累计第8次有效颠球未达到。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 14.96,
+      "log_step": 141,
+      "text_zh": "观察：视频 14.72–14.96 s 足球从高处下落至机器人右小腿附近，录像随即结束，没有后续合法离脚上升；log step 141 针对交付文件的验证回放报告 failed=\"the ball touched a part other than a foot\"、stages 仅 touch 1 至 touch 4，reward.stages_reached=4、metric_touches=4。推断：累计第9次有效颠球未达到。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 14.96,
+      "log_step": 141,
+      "text_zh": "观察：视频 14.72–14.96 s 足球从高处下落至机器人右小腿附近，录像随即结束，没有后续合法离脚上升；log step 141 针对交付文件的验证回放报告 failed=\"the ball touched a part other than a foot\"、stages 仅 touch 1 至 touch 4，reward.stages_reached=4、metric_touches=4。推断：累计第10次有效颠球未达到。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "value": 4.0,
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better"
+  },
+  {
+   "name": "ball_height_m",
+   "value": 0.354,
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_5",
+   "t": 14.96,
+   "log_step": 141,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 14.20–14.60 s 右脚击球后足球上升，视频 14.72–14.96 s 足球又落向右小腿前侧，右腿正在迎球；末帧仍站立。log step 141 对已交付文件的验证回放确认仅完成四次，因“the ball touched a part other than a foot”终止。推断：失败发生在争取第5次有效颠球的迎击中，此后没有修复。"
+  },
+  "unmet": {
+   "juggles_counted_1": true,
+   "juggles_counted_2": true,
+   "juggles_counted_3": true,
+   "juggles_counted_4": true,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 14.96,
+    "log_step": 141,
+    "evidence_zh": "观察：视频 14.72–14.96 s 足球快速下落，右腿迎球时足球到达小腿前侧，没有像视频 14.20–14.60 s 那样从合法脚部弹起；log step 141 的交付回放报告非脚部触球。推断：右脚未能在下落球到达时以合法部位迎击，属于与运动中足球的迎击配合失败。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 14.96,
+    "log_step": 140,
+    "evidence_zh": "推断：视频 12.36–14.60 s 已能连续用右脚将球击起，但视频 14.72–14.96 s 的下一次迎击没有把接触维持在合法脚部；log step 140 证实脚本重建的交付轨迹逐项相同，log step 141 的独立回放复现四次后非脚触球，支持迎球控制失败。agent 在 log step 142–143 也自述后续来球偏离落点或速度过快时会碰小腿，但其所称碰撞体尺寸和可行区域未独立验证，不能据此确定接触设计、逆解或反馈的具体子类。",
+    "source_zh": "视频 / 日志 / agent 自述"
+   }
+  ],
+  "gap_zh": "观察：inputs/ 没有本次状态日志，录像为25帧每秒且没有逐帧裁判计数。推断：四次计数与可见上升的配对依赖交付重建日志及确定性验证回放；视频 11.84–12.16 s 首次反弹为何未计数，无法独立区分高度、能量折算速度等门槛。末次非脚触球由验证回放确认，右小腿位置由视频和 agent 自述相互支持，但精确碰撞点、合法区域边界及过早/过晚或跟踪误差的细分机制不能从帧图独立量出。"
+ },
+ "summary_zh": "机器人以左脚支撑、右脚迎球，完成第1至第4次有效颠球，十个节点终态为4/10；视频 11.84–12.16 s 还有一次未计入的低反弹。第5次迎击在视频约14.96 s 因球触及非脚部而结束，机器人未跌倒，断点为第5次颠球，归因为迎球控制失败。没有状态日志及逐帧裁判读数，达标时刻取可见上升的近似时间，计数和非脚部犯规由交付文件的验证回放补足。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "四个节点已达成、第5次未达成，断点在第5次颠球（右小腿触球犯规），帧图与交付回放日志一致；未发现需要改动的内容。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:59:26Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--sfy2pjr/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--sfy2pjr/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "e62213e2e24cdbd51927e7463a4418a5f59875f56e8bcc41fad2f231a660a14a",
+   "video_sha256": "88cafa21ec8b0de751d89ec8a246406c7104b658eb9227489849f3593e37ef7e",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--uovpsef.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--uovpsef.json
new file mode 100644
index 0000000000000000000000000000000000000000..0901baa20bc7cb8d1520883ced68aa8b3ae4f29e
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--uovpsef.json
@@ -0,0 +1,318 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-privileged--uovpsef",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 372,
+   "sim_steps": 372,
+   "horizon": 6000,
+   "episode_over": 1,
+   "progress": 0.0,
+   "stages_reached": 0,
+   "stages_total": 10,
+   "fell": 0,
+   "replay_s": 7.69,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 0.302,
+   "metric_bounces": 0.0,
+   "metric_touches": 0.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 132,
+   "agent_resets": 0,
+   "agent_replays": 2,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.416,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.98,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-privileged--uovpsef/replay.mp4",
+  "duration_s": 3.76,
+  "fps": 25.0,
+  "frames": 94,
+  "sha256": "c8b6fdce221814fb612aa9a7f0527b42e93407363397a1513204c0c6888d79d4"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频初态及视频0.20–3.20 s，机器人以双掌从两侧夹住足球并调整腿部站姿；没有可见的脚部击球后离脚上升。"
+     },
+     {
+      "t": 3.72,
+      "log_step": 162,
+      "text_zh": "观察：视频3.32–3.72 s，机器人逐渐侧倾下沉，足球由腰前下落至大腿附近，未出现一次合法脚部击打后的升高。log step 162对交付文件的回放报告stages=[]、touches=0、bounces=0，并以“the ball touched a part other than a foot”结束；inputs/run.json评分字段stages_reached=0、metric_touches=0与此一致。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "观察：视频0.00–3.32 s可见两只平掌从足球两侧挤压夹持，随后放球；候选接触名“对向挤压夹持”。单只手没有包围足球，因此不标抓握；这是首次颠球尝试中的放球准备。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.00–3.72 s，机器人双脚与草地承重接触，在调整击球站姿和放球时支撑身体；末段身体侧倾下沉。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频3.72 s足球下落至机器人右腿上部并与其相贴，交付回放log step 162确认非脚部触球；候选接触名“非预期碰撞”，不是可见的脚部击打。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.72,
+      "log_step": 162,
+      "text_zh": "推断：第2次有效颠球未达到。视频0.00–3.72 s只有持球、站姿调整及末段球下落，未见连续脚部击球后升高；log step 162的交付回放与inputs/run.json评分均报告有效计数0、已达阶段0/10。首次计数未发生，因此后续累计第2次计数也未发生。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.72,
+      "log_step": 162,
+      "text_zh": "推断：第3次有效颠球未达到。视频0.00–3.72 s只有持球、站姿调整及末段球下落，未见连续脚部击球后升高；log step 162的交付回放与inputs/run.json评分均报告有效计数0、已达阶段0/10。首次计数未发生，因此后续累计第3次计数也未发生。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.72,
+      "log_step": 162,
+      "text_zh": "推断：第4次有效颠球未达到。视频0.00–3.72 s只有持球、站姿调整及末段球下落，未见连续脚部击球后升高；log step 162的交付回放与inputs/run.json评分均报告有效计数0、已达阶段0/10。首次计数未发生，因此后续累计第4次计数也未发生。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.72,
+      "log_step": 162,
+      "text_zh": "推断：第5次有效颠球未达到。视频0.00–3.72 s只有持球、站姿调整及末段球下落，未见连续脚部击球后升高；log step 162的交付回放与inputs/run.json评分均报告有效计数0、已达阶段0/10。首次计数未发生，因此后续累计第5次计数也未发生。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.72,
+      "log_step": 162,
+      "text_zh": "推断：第6次有效颠球未达到。视频0.00–3.72 s只有持球、站姿调整及末段球下落，未见连续脚部击球后升高；log step 162的交付回放与inputs/run.json评分均报告有效计数0、已达阶段0/10。首次计数未发生，因此后续累计第6次计数也未发生。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.72,
+      "log_step": 162,
+      "text_zh": "推断：第7次有效颠球未达到。视频0.00–3.72 s只有持球、站姿调整及末段球下落，未见连续脚部击球后升高；log step 162的交付回放与inputs/run.json评分均报告有效计数0、已达阶段0/10。首次计数未发生，因此后续累计第7次计数也未发生。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.72,
+      "log_step": 162,
+      "text_zh": "推断：第8次有效颠球未达到。视频0.00–3.72 s只有持球、站姿调整及末段球下落，未见连续脚部击球后升高；log step 162的交付回放与inputs/run.json评分均报告有效计数0、已达阶段0/10。首次计数未发生，因此后续累计第8次计数也未发生。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.72,
+      "log_step": 162,
+      "text_zh": "推断：第9次有效颠球未达到。视频0.00–3.72 s只有持球、站姿调整及末段球下落，未见连续脚部击球后升高；log step 162的交付回放与inputs/run.json评分均报告有效计数0、已达阶段0/10。首次计数未发生，因此后续累计第9次计数也未发生。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.72,
+      "log_step": 162,
+      "text_zh": "推断：第10次有效颠球未达到。视频0.00–3.72 s只有持球、站姿调整及末段球下落，未见连续脚部击球后升高；log step 162的交付回放与inputs/run.json评分均报告有效计数0、已达阶段0/10。首次计数未发生，因此后续累计第10次计数也未发生。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better",
+   "value": 0.0
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": 0.302
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_1",
+   "t": 3.4,
+   "log_step": 162,
+   "where": "操作中",
+   "evidence_zh": "观察：视频3.32–3.72 s，机器人正在放球及尝试首次颠球，身体持续向画面左侧侧倾下沉，球落向腿部，直到末帧仍未恢复直立或产生一次离脚上升；log step 162的交付回放确认非脚部触球终止、0/10阶段。log step 190再次生成交付文件所得校验和不变，说明后续会话测试没有替换这段交付。断点取首次颠球的失稳过程，非脚部触球是此步的终止事件。"
+  },
+  "unmet": {
+   "juggles_counted_1": false,
+   "juggles_counted_2": false,
+   "juggles_counted_3": false,
+   "juggles_counted_4": false,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 3.6,
+    "log_step": null,
+    "evidence_zh": "观察：视频3.32–3.72 s逐帧可见机器人侧倾不断加大，躯干下沉、双腿斜伸，至末帧没有恢复稳定站立；足球同时落至腿部附近，未被脚向上颠起。这里标可见的全身失稳；inputs/run.json的fell=0表示评分器在非脚部触球终止前尚未判定跌倒，不把会话中其他试验的摔倒算入本轨迹。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：站姿变换及放球阶段的全身姿态控制未能稳住机器人。log step 132记录交付生成器对右腿关节作插值后保持目标，以骨盆偏航反馈修正左腿逆解并施加支撑前馈力矩；log step 162、190确认该生成器交付的是相同372行动的可重复失败轨迹。视频0.60–3.32 s可见机器人逐渐转身，视频3.32–3.72 s进一步侧倾下沉，表明实际全身运动没有保持生成器所需的稳定击球站姿。证据不足以进一步断定是接触布局、逆解目标、反馈律还是某个增益参数所致，故停在“控制”。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志，固定远景不能精确分辨末段接触点及毫秒接触顺序；非脚部触球由交付文件回放结果确认，未凭会话测试的具体身体部位替代末帧证据。缺少全身受力和姿态跟踪诊断，无法可靠细分控制根因；未将25 fps视频用于估算0.05秒接触或0.95倍速度门槛。"
+ },
+ "summary_zh": "机器人完成持球状态下的站姿调整并开始放球，但首次有效颠球未发生，十个计数节点均未达到，评分器有效触球和反弹计数均为0。视频3.32–3.72 s显示机器人持续侧倾下沉、球落向腿部，交付回放以非脚部触球结束；日志中的生成器与可重复回放支持将根因保守归为控制。缺少逐帧状态和受力数据，无法精确定位碰撞点或细分失稳的控制原因。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "十个节点均未达到与评分器（stages_reached=0、touches=0）一致；视频3.2–3.7 s可见机器人侧倾下沉、球仍在腰腹处后落向腿部，断点、平衡现象和控制根因有证据支持，未发现需修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:54:33Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--uovpsef/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--uovpsef/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "3335acbeaa990c68599b4ea15bbd4b3bc6f8d93af956aa799dae3e74f4bae1f0",
+   "video_sha256": "c8b6fdce221814fb612aa9a7f0527b42e93407363397a1513204c0c6888d79d4",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--x6wdyax.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--x6wdyax.json
new file mode 100644
index 0000000000000000000000000000000000000000..44bc19bd98376af19bb25600a54f1ea2e9f7a62b
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--x6wdyax.json
@@ -0,0 +1,314 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-privileged--x6wdyax",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 573,
+   "sim_steps": 573,
+   "horizon": 6000,
+   "episode_over": 1,
+   "progress": 0.0,
+   "stages_reached": 0,
+   "stages_total": 10,
+   "fell": 1,
+   "replay_s": 3.57,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 0.208,
+   "metric_bounces": 0.0,
+   "metric_touches": 0.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 60,
+   "agent_resets": 89,
+   "agent_replays": 0,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.496,
+   "peak_processes": 11,
+   "peak_disk_gb": 0.242,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-privileged--x6wdyax/replay.mp4",
+  "duration_s": 5.76,
+  "fps": 25.0,
+  "frames": 144,
+  "sha256": "df24ddb2ea0cdda1dd5a37189fe08e1b6568207d3dc19f27be7ac036955f37fb"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.8,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–4.80 s，机器人双掌持续挤持足球，并转动上身、调整腿部姿态；球没有从脚部击打后离脚上升。"
+     },
+     {
+      "t": 5.72,
+      "log_step": 34,
+      "text_zh": "观察：视频 5.32–5.72 s，机器人持球倾倒、骨盆快速降低，足球随双手下降，未出现一次离脚上升的颠球；log step 34 保存的轨迹为573步、fell、touches=0，与评分器 stages_reached=0、metric_touches=0、fell=1一致。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "观察：视频 0.00–5.72 s，机器人两掌从两侧挤持足球，并在调整姿态时一起带动球；候选接触名“双掌挤持”，两只平掌均未包围足球，不属于抓握，现有方式没有对应项。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–5.32 s，机器人双脚接触草地承重，并在首次颠球的准备中调整腿部姿态；视频 5.32–5.72 s 支撑失稳。未见脚对足球完成击打。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.72,
+      "log_step": null,
+      "text_zh": "推断：视频 0.00–5.72 s 的足球一直随双掌移动，末段随机器人倒下，未见连续离脚上升；评分器 stages_reached=0、metric_touches=0。第1次有效颠球未达成，故依次计数的第2次颠球亦未达成，机器人尚未进入该次尝试。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.72,
+      "log_step": null,
+      "text_zh": "推断：视频 0.00–5.72 s 的足球一直随双掌移动，末段随机器人倒下，未见连续离脚上升；评分器 stages_reached=0、metric_touches=0。第1次有效颠球未达成，故依次计数的第3次颠球亦未达成，机器人尚未进入该次尝试。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.72,
+      "log_step": null,
+      "text_zh": "推断：视频 0.00–5.72 s 的足球一直随双掌移动，末段随机器人倒下，未见连续离脚上升；评分器 stages_reached=0、metric_touches=0。第1次有效颠球未达成，故依次计数的第4次颠球亦未达成，机器人尚未进入该次尝试。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.72,
+      "log_step": null,
+      "text_zh": "推断：视频 0.00–5.72 s 的足球一直随双掌移动，末段随机器人倒下，未见连续离脚上升；评分器 stages_reached=0、metric_touches=0。第1次有效颠球未达成，故依次计数的第5次颠球亦未达成，机器人尚未进入该次尝试。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.72,
+      "log_step": null,
+      "text_zh": "推断：视频 0.00–5.72 s 的足球一直随双掌移动，末段随机器人倒下，未见连续离脚上升；评分器 stages_reached=0、metric_touches=0。第1次有效颠球未达成，故依次计数的第6次颠球亦未达成，机器人尚未进入该次尝试。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.72,
+      "log_step": null,
+      "text_zh": "推断：视频 0.00–5.72 s 的足球一直随双掌移动，末段随机器人倒下，未见连续离脚上升；评分器 stages_reached=0、metric_touches=0。第1次有效颠球未达成，故依次计数的第7次颠球亦未达成，机器人尚未进入该次尝试。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.72,
+      "log_step": null,
+      "text_zh": "推断：视频 0.00–5.72 s 的足球一直随双掌移动，末段随机器人倒下，未见连续离脚上升；评分器 stages_reached=0、metric_touches=0。第1次有效颠球未达成，故依次计数的第8次颠球亦未达成，机器人尚未进入该次尝试。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.72,
+      "log_step": null,
+      "text_zh": "推断：视频 0.00–5.72 s 的足球一直随双掌移动，末段随机器人倒下，未见连续离脚上升；评分器 stages_reached=0、metric_touches=0。第1次有效颠球未达成，故依次计数的第9次颠球亦未达成，机器人尚未进入该次尝试。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 5.72,
+      "log_step": null,
+      "text_zh": "推断：视频 0.00–5.72 s 的足球一直随双掌移动，末段随机器人倒下，未见连续离脚上升；评分器 stages_reached=0、metric_touches=0。第1次有效颠球未达成，故依次计数的第10次颠球亦未达成，机器人尚未进入该次尝试。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "value": 0.0,
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better"
+  },
+  {
+   "name": "ball_height_m",
+   "value": 0.208,
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_1",
+   "t": 5.3,
+   "log_step": 34,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 5.32–5.72 s，机器人在持球调整腿部、准备首次颠球时身体持续倾倒、骨盆降近草地，未恢复站立，也未完成脚部击球。log step 34 明确保存573步、fell、touches=0的轨迹，评分器同样记录 fell=1、stages_reached=0；失败位于第1次颠球的准备操作中。"
+  },
+  "unmet": {
+   "juggles_counted_1": false,
+   "juggles_counted_2": false,
+   "juggles_counted_3": false,
+   "juggles_counted_4": false,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 5.3,
+    "log_step": null,
+    "evidence_zh": "观察：视频 5.32–5.72 s，机器人持球向前侧方倾倒，双腿弯折，骨盆快速下降，末帧已接近草地且未恢复；评分器 fell=1。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "log_step": 34,
+    "evidence_zh": "推断：视频 4.72–5.72 s 显示机器人在调整腿部支撑时姿态不断偏离直立并倒下；log step 21 的同一run2方案在准备右脚悬停时报告失衡，log step 34 随后以该方案保存573步失败轨迹，log step 95 的交付说明确认保留此早期轨迹。首次脚触球之前的全身动作未得到稳定控制，导致本次失败；不足以区分轨迹设计、反馈跟踪或具体参数问题。",
+    "source_zh": "视频 / 日志"
+   }
+  ],
+  "gap_zh": "没有状态日志，固定场边视频不能确定精确接触力、重心和关节跟踪误差，故控制根因不细分。视频未显示脚部击球或离脚上升，裁判有效计数也为0，这些缺口不影响十个节点均未达成的判断。后续会话调试未覆盖log step 34保存的交付文件；节点及现象仅依据本次回放，不采用后续测试结果。"
+ },
+ "summary_zh": "机器人进行了持球转身和首次颠球的姿态准备，但有效颠球为0次，十个节点均未达成。视频约 5.3–5.72 s，机器人仍双掌夹球便失衡倒下，断在第1次颠球操作中，根因可归于全身控制未稳定。缺少状态日志，无法进一步区分具体轨迹、反馈或参数问题；交付为log step 34保存的早期失败轨迹。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "十个节点均未达成、断点在第1次颠球（平衡/全身倒下）、根因停在控制均有视频与评分器 fell=1、touches=0 支持；未发现需修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:57:14Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--x6wdyax/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--x6wdyax/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "0a348762112ce5d8c7b081a3e248fe2d075c78f48e1aa6dd6f8fe6ee65cc2f60",
+   "video_sha256": "df24ddb2ea0cdda1dd5a37189fe08e1b6568207d3dc19f27be7ac036955f37fb",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--zyxoa98.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--zyxoa98.json
new file mode 100644
index 0000000000000000000000000000000000000000..0aaff36f07a62fbb82f4700e4f000474e7ec16ff
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-privileged--zyxoa98.json
@@ -0,0 +1,488 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-privileged--zyxoa98",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 926,
+   "sim_steps": 926,
+   "horizon": 6000,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 10,
+   "stages_total": 10,
+   "fell": 0,
+   "replay_s": 5.43,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 0.508,
+   "metric_bounces": 0.0,
+   "metric_touches": 10.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 79,
+   "agent_resets": 21,
+   "agent_replays": 2,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.767,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.055,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-privileged--zyxoa98/replay.mp4",
+  "duration_s": 9.28,
+  "fps": 25.0,
+  "frames": 232,
+  "sha256": "71ef3c97e05ef20d7bbbb4fac5b078b28d5a2828bef415e187cba0571d5f64e5"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 10,
+  "final": 10,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.0,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.64,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00 s 足球位于机器人两掌之间；视频 2.40–2.52 s 仍由双掌夹持，视频 2.64–2.84 s 两掌分开，足球离开双手向左脚下落。放球属于首次颠球的准备动作。"
+     },
+     {
+      "t": 3.0,
+      "log_step": null,
+      "text_zh": "观察：视频 2.84–2.92 s，足球下落至机器人左脚（画面右侧前伸脚）后转为上升；至视频 3.12 s，球已离脚，机器人右脚仍支撑草地。这是放球后按视频顺序辨认的第1次击球。"
+     },
+     {
+      "t": 3.0,
+      "log_step": null,
+      "text_zh": "推断：以视频 3.00 s 附近的离脚上升作为第1次有效颠球达到高度条件的近似时刻，误差约0.1 s；评分器 reward.stages_reached=10、metric_touches=10、success=1 支持该次最终计入。25帧每秒画面不能独立精测接触历时或0.95倍离球速度门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "观察：视频 0.00–2.52 s，机器人两侧平掌共同挤压并移动足球，随后分开双掌放球；候选名为“双掌挤压夹持”，两手均未包合足球，不标抓握；此接触先于首次脚部击打。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 2.84–3.12 s，机器人前伸的左脚向上击打足球，球随后离脚上升；候选接触名为“击打”，现有方式没有击打项，推/扫或托/舀不能准确描述这次短促踢击。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.84–3.12 s，机器人右脚接触草地并承重，支撑左脚颠球时的身体。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.72,
+      "log_step": null,
+      "text_zh": "观察：视频 3.56–3.64 s，足球下落至机器人左脚（画面右侧前伸脚）后转为上升；至视频 3.84 s，球已离脚，机器人右脚仍支撑草地。这是放球后按视频顺序辨认的第2次击球。"
+     },
+     {
+      "t": 3.72,
+      "log_step": null,
+      "text_zh": "推断：以视频 3.72 s 附近的离脚上升作为第2次有效颠球达到高度条件的近似时刻，误差约0.1 s；评分器 reward.stages_reached=10、metric_touches=10、success=1 支持该次最终计入。25帧每秒画面不能独立精测接触历时或0.95倍离球速度门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 3.56–3.84 s，机器人前伸的左脚向上击打足球，球随后离脚上升；候选接触名为“击打”，现有方式没有击打项，推/扫或托/舀不能准确描述这次短促踢击。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 3.56–3.84 s，机器人右脚接触草地并承重，支撑左脚颠球时的身体。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.4,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.4,
+      "log_step": null,
+      "text_zh": "观察：视频 4.28–4.36 s，足球下落至机器人左脚（画面右侧前伸脚）后转为上升；至视频 4.56 s，球已离脚，机器人右脚仍支撑草地。这是放球后按视频顺序辨认的第3次击球。"
+     },
+     {
+      "t": 4.4,
+      "log_step": null,
+      "text_zh": "推断：以视频 4.40 s 附近的离脚上升作为第3次有效颠球达到高度条件的近似时刻，误差约0.1 s；评分器 reward.stages_reached=10、metric_touches=10、success=1 支持该次最终计入。25帧每秒画面不能独立精测接触历时或0.95倍离球速度门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 4.28–4.56 s，机器人前伸的左脚向上击打足球，球随后离脚上升；候选接触名为“击打”，现有方式没有击打项，推/扫或托/舀不能准确描述这次短促踢击。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 4.28–4.56 s，机器人右脚接触草地并承重，支撑左脚颠球时的身体。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.12,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.12,
+      "log_step": null,
+      "text_zh": "观察：视频 5.00–5.08 s，足球下落至机器人左脚（画面右侧前伸脚）后转为上升；至视频 5.28 s，球已离脚，机器人右脚仍支撑草地。这是放球后按视频顺序辨认的第4次击球。"
+     },
+     {
+      "t": 5.12,
+      "log_step": null,
+      "text_zh": "推断：以视频 5.12 s 附近的离脚上升作为第4次有效颠球达到高度条件的近似时刻，误差约0.1 s；评分器 reward.stages_reached=10、metric_touches=10、success=1 支持该次最终计入。25帧每秒画面不能独立精测接触历时或0.95倍离球速度门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 5.00–5.28 s，机器人前伸的左脚向上击打足球，球随后离脚上升；候选接触名为“击打”，现有方式没有击打项，推/扫或托/舀不能准确描述这次短促踢击。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 5.00–5.28 s，机器人右脚接触草地并承重，支撑左脚颠球时的身体。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.8,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.8,
+      "log_step": null,
+      "text_zh": "观察：视频 5.64–5.72 s，足球下落至机器人左脚（画面右侧前伸脚）后转为上升；至视频 5.96 s，球已离脚，机器人右脚仍支撑草地。这是放球后按视频顺序辨认的第5次击球。"
+     },
+     {
+      "t": 5.8,
+      "log_step": null,
+      "text_zh": "推断：以视频 5.80 s 附近的离脚上升作为第5次有效颠球达到高度条件的近似时刻，误差约0.1 s；评分器 reward.stages_reached=10、metric_touches=10、success=1 支持该次最终计入。25帧每秒画面不能独立精测接触历时或0.95倍离球速度门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 5.64–5.96 s，机器人前伸的左脚向上击打足球，球随后离脚上升；候选接触名为“击打”，现有方式没有击打项，推/扫或托/舀不能准确描述这次短促踢击。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 5.64–5.96 s，机器人右脚接触草地并承重，支撑左脚颠球时的身体。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.48,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.48,
+      "log_step": null,
+      "text_zh": "观察：视频 6.36–6.44 s，足球下落至机器人左脚（画面右侧前伸脚）后转为上升；至视频 6.68 s，球已离脚，机器人右脚仍支撑草地。这是放球后按视频顺序辨认的第6次击球。"
+     },
+     {
+      "t": 6.48,
+      "log_step": null,
+      "text_zh": "推断：以视频 6.48 s 附近的离脚上升作为第6次有效颠球达到高度条件的近似时刻，误差约0.1 s；评分器 reward.stages_reached=10、metric_touches=10、success=1 支持该次最终计入。25帧每秒画面不能独立精测接触历时或0.95倍离球速度门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 6.36–6.68 s，机器人前伸的左脚向上击打足球，球随后离脚上升；候选接触名为“击打”，现有方式没有击打项，推/扫或托/舀不能准确描述这次短促踢击。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 6.36–6.68 s，机器人右脚接触草地并承重，支撑左脚颠球时的身体。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.2,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.2,
+      "log_step": null,
+      "text_zh": "观察：视频 7.04–7.12 s，足球下落至机器人左脚（画面右侧前伸脚）后转为上升；至视频 7.36 s，球已离脚，机器人右脚仍支撑草地。这是放球后按视频顺序辨认的第7次击球。"
+     },
+     {
+      "t": 7.2,
+      "log_step": null,
+      "text_zh": "推断：以视频 7.20 s 附近的离脚上升作为第7次有效颠球达到高度条件的近似时刻，误差约0.1 s；评分器 reward.stages_reached=10、metric_touches=10、success=1 支持该次最终计入。25帧每秒画面不能独立精测接触历时或0.95倍离球速度门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 7.04–7.36 s，机器人前伸的左脚向上击打足球，球随后离脚上升；候选接触名为“击打”，现有方式没有击打项，推/扫或托/舀不能准确描述这次短促踢击。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 7.04–7.36 s，机器人右脚接触草地并承重，支撑左脚颠球时的身体。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.92,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.92,
+      "log_step": null,
+      "text_zh": "观察：视频 7.76–7.84 s，足球下落至机器人左脚（画面右侧前伸脚）后转为上升；至视频 8.08 s，球已离脚，机器人右脚仍支撑草地。这是放球后按视频顺序辨认的第8次击球。"
+     },
+     {
+      "t": 7.92,
+      "log_step": null,
+      "text_zh": "推断：以视频 7.92 s 附近的离脚上升作为第8次有效颠球达到高度条件的近似时刻，误差约0.1 s；评分器 reward.stages_reached=10、metric_touches=10、success=1 支持该次最终计入。25帧每秒画面不能独立精测接触历时或0.95倍离球速度门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 7.76–8.08 s，机器人前伸的左脚向上击打足球，球随后离脚上升；候选接触名为“击打”，现有方式没有击打项，推/扫或托/舀不能准确描述这次短促踢击。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 7.76–8.08 s，机器人右脚接触草地并承重，支撑左脚颠球时的身体。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.6,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.6,
+      "log_step": null,
+      "text_zh": "观察：视频 8.44–8.52 s，足球下落至机器人左脚（画面右侧前伸脚）后转为上升；至视频 8.76 s，球已离脚，机器人右脚仍支撑草地。这是放球后按视频顺序辨认的第9次击球。"
+     },
+     {
+      "t": 8.6,
+      "log_step": null,
+      "text_zh": "推断：以视频 8.60 s 附近的离脚上升作为第9次有效颠球达到高度条件的近似时刻，误差约0.1 s；评分器 reward.stages_reached=10、metric_touches=10、success=1 支持该次最终计入。25帧每秒画面不能独立精测接触历时或0.95倍离球速度门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 8.44–8.76 s，机器人前伸的左脚向上击打足球，球随后离脚上升；候选接触名为“击打”，现有方式没有击打项，推/扫或托/舀不能准确描述这次短促踢击。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 8.44–8.76 s，机器人右脚接触草地并承重，支撑左脚颠球时的身体。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.24,
+      "log_step": null,
+      "text_zh": "观察：视频 9.16–9.20 s，足球下落至机器人左脚（画面右侧前伸脚）后转为上升；至视频 9.24 s，球已离脚，机器人右脚仍支撑草地。这是放球后按视频顺序辨认的第10次击球。"
+     },
+     {
+      "t": 9.24,
+      "log_step": null,
+      "text_zh": "推断：视频 9.16–9.24 s 可见最后一次左脚击球及离脚上升，末帧仍在上升；结合评分器 reward.stages_reached=10、metric_touches=10、replay_success=1，确认第10次最终计入。视频 9.24 s 是末段近似导航锚点，精确达到高度及速度条件的时刻在末帧附近，误差约0.1 s，不能由25帧每秒录像单独解析。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 9.16–9.24 s，机器人前伸的左脚向上击打足球，球随后离脚上升；候选接触名为“击打”，现有方式没有击打项，推/扫或托/舀不能准确描述这次短促踢击。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 9.16–9.24 s，机器人右脚接触草地并承重，支撑左脚颠球时的身体。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better",
+   "value": 10.0
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": 0.508
+  }
+ ],
+ "failure": null,
+ "summary_zh": "观察：机器人先将足球保持在双掌之间并准备单脚支撑，随后放球，用左脚连续完成十次可见的短促颠球，右脚持续支撑；视频 9.24 s 末帧足球正从最后一次击球后上升，未见跌倒或球落地，评分器记录有效颠球10次、成功且确定性回放通过。推断：十个事件节点均已完成，时间按逐帧画面给近似锚点；本次没有状态日志，25帧每秒录像无法精确验证毫秒级接触、离球速度和最后一次高度门槛的瞬时值，这些验收由评分器成功及计数补足，且没有独立于本次运行的参考视频。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "十个事件节点均达成，与视频和评分器（stages_reached=10、touches=10、fell=0）一致；时间为近似锚点，contact_actual（双臂协作其他、腿/躯干击打、支撑）合理，failure 为空正确。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:46:20Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--zyxoa98/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-privileged--zyxoa98/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "6b81470f118f4fdfa8876d4a47084cfd8b6cec9ecca7e5944a98c94ec4bc2e44",
+   "video_sha256": "71ef3c97e05ef20d7bbbb4fac5b078b28d5a2828bef415e187cba0571d5f64e5",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--dwfhh6g.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--dwfhh6g.json
new file mode 100644
index 0000000000000000000000000000000000000000..b5437f166e9d1e08e798492f3bcbed96e853929a
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--dwfhh6g.json
@@ -0,0 +1,311 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-standard--dwfhh6g",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 263,
+   "sim_steps": 3129,
+   "calls": 44,
+   "rows": 3129,
+   "resets": 22,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 3282.6,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.238,
+   "peak_processes": 13,
+   "peak_disk_gb": 0.481,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-standard--dwfhh6g/replay.mp4",
+  "duration_s": 2.64,
+  "fps": 25.0,
+  "frames": 66,
+  "sha256": "8bc1d4ebbf66147bf82966eee59d3d49f530ff4885082cb0dbc51d7bf471b05a"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频0.00 s机器人双掌挤压夹持足球，双脚站在草地上；视频0.48–2.20 s机器人屈膝并抬起一脚作准备，足球仍随双手运动。"
+     },
+     {
+      "t": 2.6,
+      "log_step": 181,
+      "text_zh": "观察：视频2.24–2.60 s机器人逐渐失去站立平衡，躯干倾斜、身体下沉，足球一直位于双掌之间，没有放落至脚部或离脚上升。log step 181最后一次实时提交返回done=true、outcome=fell、sim_time_s=2.63；与本段视频末尾失衡相符，第1次有效颠球未发生。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "观察：视频0.00–2.60 s两只无可动手指的平掌从两侧挤压夹持足球并随身体调整移动；候选接触名“双掌挤压夹持”，没有单手包围或夹爪闭合，故不用抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.00–2.60 s双脚先共同承重，随后机器人抬起一脚，另一脚接触草地支撑身体；准备阶段最终失衡，没有观察到脚与足球接触。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.6,
+      "log_step": null,
+      "text_zh": "观察：检查本次视频0.00–2.60 s，足球始终在机器人双掌之间，未出现首次放球及脚部击球；视频2.24–2.60 s机器人失衡而结束，未进入第2次颠球。推断：第1次颠球未完成且没有任何后续颠球，第2次计数节点未达到。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.6,
+      "log_step": null,
+      "text_zh": "观察：检查本次视频0.00–2.60 s，足球始终在机器人双掌之间，未出现首次放球及脚部击球；视频2.24–2.60 s机器人失衡而结束，未进入第3次颠球。推断：第1次颠球未完成且没有任何后续颠球，第3次计数节点未达到。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.6,
+      "log_step": null,
+      "text_zh": "观察：检查本次视频0.00–2.60 s，足球始终在机器人双掌之间，未出现首次放球及脚部击球；视频2.24–2.60 s机器人失衡而结束，未进入第4次颠球。推断：第1次颠球未完成且没有任何后续颠球，第4次计数节点未达到。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.6,
+      "log_step": null,
+      "text_zh": "观察：检查本次视频0.00–2.60 s，足球始终在机器人双掌之间，未出现首次放球及脚部击球；视频2.24–2.60 s机器人失衡而结束，未进入第5次颠球。推断：第1次颠球未完成且没有任何后续颠球，第5次计数节点未达到。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.6,
+      "log_step": null,
+      "text_zh": "观察：检查本次视频0.00–2.60 s，足球始终在机器人双掌之间，未出现首次放球及脚部击球；视频2.24–2.60 s机器人失衡而结束，未进入第6次颠球。推断：第1次颠球未完成且没有任何后续颠球，第6次计数节点未达到。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.6,
+      "log_step": null,
+      "text_zh": "观察：检查本次视频0.00–2.60 s，足球始终在机器人双掌之间，未出现首次放球及脚部击球；视频2.24–2.60 s机器人失衡而结束，未进入第7次颠球。推断：第1次颠球未完成且没有任何后续颠球，第7次计数节点未达到。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.6,
+      "log_step": null,
+      "text_zh": "观察：检查本次视频0.00–2.60 s，足球始终在机器人双掌之间，未出现首次放球及脚部击球；视频2.24–2.60 s机器人失衡而结束，未进入第8次颠球。推断：第1次颠球未完成且没有任何后续颠球，第8次计数节点未达到。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.6,
+      "log_step": null,
+      "text_zh": "观察：检查本次视频0.00–2.60 s，足球始终在机器人双掌之间，未出现首次放球及脚部击球；视频2.24–2.60 s机器人失衡而结束，未进入第9次颠球。推断：第1次颠球未完成且没有任何后续颠球，第9次计数节点未达到。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.6,
+      "log_step": null,
+      "text_zh": "观察：检查本次视频0.00–2.60 s，足球始终在机器人双掌之间，未出现首次放球及脚部击球；视频2.24–2.60 s机器人失衡而结束，未进入第10次颠球。推断：第1次颠球未完成且没有任何后续颠球，第10次计数节点未达到。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_1",
+   "t": 2.4,
+   "log_step": 181,
+   "where": "操作中",
+   "evidence_zh": "观察：视频0.48–2.20 s机器人持球屈膝、移重并抬脚，为首次颠球调整支撑；视频2.24–2.60 s身体逐渐倾斜并下沉，球仍在双掌间，未放落、未踢球，也未恢复站稳。log step 181最后一次实时提交由机器人服务返回outcome=fell；之后log step 182–185只有说明及文件操作，没有新回合。推断：失败发生在第1次颠球的准备动作中，此步未被修复。"
+  },
+  "unmet": {
+   "juggles_counted_1": false,
+   "juggles_counted_2": false,
+   "juggles_counted_3": false,
+   "juggles_counted_4": false,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 2.4,
+    "log_step": 181,
+    "evidence_zh": "观察：视频2.16–2.60 s机器人一脚抬起，身体持续倾斜、下沉，支撑姿态不能维持；足球仍在手边而没有完成脚部击打。log step 181机器人服务对最后一次实时回合返回done=true、reason=failed、outcome=fell。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：log step 124的控制脚本先按预定骨盆与踝部路径计算整段关节目标，再批量提交lowcmd；log step 179、181由该脚本派生并运行最终att5，最后一次提交在放球前即返回fell。log step 182的说明中agent自述没有尝试姿态反馈修正回路。结合视频2.24–2.60 s抬脚准备时的持续失稳，说明当前关节轨迹执行和支撑控制未能维持动态平衡；证据不足以进一步判定为逆解、承载布局或某个增益参数错误。",
+    "source_zh": "日志"
+   }
+  ],
+  "gap_zh": "观察：inputs/有本次视频、完整run.json及成功参考视频，但没有状态日志；reward没有报告touches或ball_height_m，故两项原生指标保留null。视频末帧为2.60 s，显示失衡过程而未完整展示倒地后的姿态，终止判定由log step 181机器人服务的fell结果补充。推断：无本次连续姿态、接触力和重心数据，不能把控制根因进一步细分；早期测试的逆解错误或agent关于向后跌倒的解释不直接当作本次的已证实原因。"
+ },
+ "summary_zh": "观察：本次视频0.00–2.60 s中，机器人持球屈膝移重并抬脚准备，足球始终在双掌间，十次有效颠球均未完成，最终进度为0/10。观察：视频2.24–2.60 s机器人失衡，log step 181最后一次实时回合返回fell；推断：断点为第1次颠球的准备动作，当前控制未能维持支撑平衡。观察：没有状态日志或裁判颠球计数、球高指标，控制的具体失稳机制无法进一步确定。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频0–2.6 s球始终在双掌间、机器人抬脚后失衡，十个节点均未达成，断点与平衡现象、控制根因均有证据支持。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:57:23Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--dwfhh6g/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--dwfhh6g/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "68a7baada2268b27b6ab33ab385fd2f4d578b10387a2a0619a0b5295867f2ac4",
+   "video_sha256": "8bc1d4ebbf66147bf82966eee59d3d49f530ff4885082cb0dbc51d7bf471b05a",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--nkero7w.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--nkero7w.json
new file mode 100644
index 0000000000000000000000000000000000000000..ff11ecd1d2fe5553438b1721bb51e524377f5ea4
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--nkero7w.json
@@ -0,0 +1,342 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-standard--nkero7w",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 748,
+   "sim_steps": 32852,
+   "calls": 15174,
+   "rows": 32852,
+   "resets": 50,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 3770.1,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.262,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.365,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-standard--nkero7w/replay.mp4",
+  "duration_s": 7.52,
+  "fps": 25.0,
+  "frames": 188,
+  "sha256": "c98fc1edcdedc59553058e41a2b3b7424234210fdecd9b0702e2851e780aa4f2"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 2,
+  "final": 2,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.04,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.04,
+      "log_step": 124,
+      "text_zh": "观察：视频5.56–5.88 s机器人分开双掌，足球落向前置右脚；视频5.92 s球在右脚前端附近转为上升，视频6.04–6.28 s已离脚升至膝部以上。log step 124最终回合的服务返回stages=2，支持本次击球被计入；完成锚点取视频6.04 s，约有0.1 s误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "观察：视频0.00–5.52 s两块平掌垫从两侧挤压足球，视频5.56–5.64 s分开使球下落；候选接触名为“双掌挤压夹持”，两只平掌各自没有包围或夹爪式合拢物体，故不标抓握。此为首次颠球的准备接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频5.92 s前置右脚前端将下落足球击起；候选接触名为“击打”，现有接触方式没有对应项。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.00–6.28 s机器人以双脚接触草地承重，准备时挪动右脚形成前后站姿，击球时左脚维持支撑。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.8,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.8,
+      "log_step": 124,
+      "text_zh": "观察：视频6.32–6.64 s足球回落，视频6.68 s在右脚前端附近再次改变运动方向，视频6.72–7.04 s离脚向上并向前飞；视频6.80 s已明显升过膝部。log step 124最终回合服务返回stages=2，与两次上升一致；完成锚点取视频6.80 s，约有0.1 s误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频6.68 s机器人右脚前端再次击打回落足球；候选接触名为“击打”，无对应的既有接触方式。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频6.32–7.04 s机器人保持前后站姿，双脚接触草地承重，左脚在右脚上翘击球时维持支撑。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.36,
+      "log_step": 124,
+      "text_zh": "观察：视频6.80–7.48 s第二次击出的足球持续向机器人前方偏移，视频7.24–7.48 s在双脚前方落向草地；机器人保持原站位，右脚虽有上翘动作，未出现第三次球脚接触或新的离脚上升。log step 124最终回合服务返回stages=2及“the ball touched the ground”，log step 125确认该回合episode=51已失败。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频7.04–7.48 s机器人双脚继续接触草地承重，右脚上翘但未跟到前移足球下方；没有观察到本次续颠所需的球脚接触，因此不记第三次击打接触。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 124,
+      "text_zh": "观察：视频5.56–7.48 s仅见两次脚部击球后的上升，随后球飞离脚前并落地；log step 124最终回合服务返回stages=2、球触地结束，未发生第4次有效颠球。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 124,
+      "text_zh": "观察：视频5.56–7.48 s仅见两次脚部击球后的上升，随后球飞离脚前并落地；log step 124最终回合服务返回stages=2、球触地结束，未发生第5次有效颠球。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 124,
+      "text_zh": "观察：视频5.56–7.48 s仅见两次脚部击球后的上升，随后球飞离脚前并落地；log step 124最终回合服务返回stages=2、球触地结束，未发生第6次有效颠球。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 124,
+      "text_zh": "观察：视频5.56–7.48 s仅见两次脚部击球后的上升，随后球飞离脚前并落地；log step 124最终回合服务返回stages=2、球触地结束，未发生第7次有效颠球。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 124,
+      "text_zh": "观察：视频5.56–7.48 s仅见两次脚部击球后的上升，随后球飞离脚前并落地；log step 124最终回合服务返回stages=2、球触地结束，未发生第8次有效颠球。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 124,
+      "text_zh": "观察：视频5.56–7.48 s仅见两次脚部击球后的上升，随后球飞离脚前并落地；log step 124最终回合服务返回stages=2、球触地结束，未发生第9次有效颠球。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 124,
+      "text_zh": "观察：视频5.56–7.48 s仅见两次脚部击球后的上升，随后球飞离脚前并落地；log step 124最终回合服务返回stages=2、球触地结束，未发生第10次有效颠球。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_3",
+   "t": 7.36,
+   "log_step": 124,
+   "where": "操作中",
+   "evidence_zh": "观察：视频6.68 s第二次击球后，足球向前飞离；视频7.24–7.48 s机器人没有移脚追到落点下方，第三次续颠未接触球，球落向草地。log step 124明确执行最终配置，服务返回stages=2、球触地；log step 125确认episode=51结束。断点为正在续接的第3次颠球，前两次已完成。"
+  },
+  "unmet": {
+   "juggles_counted_1": true,
+   "juggles_counted_2": true,
+   "juggles_counted_3": false,
+   "juggles_counted_4": false,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 7.36,
+    "log_step": 124,
+    "evidence_zh": "观察：视频6.80–7.48 s足球上升并向前飞出后回落，机器人仍保持前后站姿，右脚的上翘没有跟上移动中的足球，球从脚前方落下，未形成第三次触球；log step 124最终回合以球触地结束。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 7.36,
+    "log_step": 124,
+    "evidence_zh": "推断：视频6.80–7.48 s第二次触球后的前向偏移没有被脚部移动补偿，是续颠控制失败。log step 71定义按预设时刻上翘右踝的击球函数，log step 97、102将其接入根据本体姿态和关节读数修正重心的循环；log step 124最终配置仍使用三组预设踝部上翘时刻并返回仅2次有效计数。日志与视频共同支持控制未能把后续击球位置接到球路上，但不能唯一确定是接触几何、轨迹跟踪还是具体增益所致，故保留“控制”总项。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "没有本次状态日志或逐帧裁判计数；25 fps视频不能独立核验0.05秒接触簇、精确速度及高度阈值，前两次有效性由最终回合服务stages=2与视频共同支持，节点时间为视觉近似。未观察到足够证据验证agent在log step 126所称胫部碰撞几何及毫秒级敏感性，不据此细分根因。reward字段未报告touches及ball_height_m，两个native_metrics值置null；服务最终回合的stages=2另作节点证据。"
+ },
+ "summary_zh": "评分视频显示机器人分开双掌放球，并以右脚完成两次有效颠球，分别在视频约6.04 s、6.80 s达到上升判据，最终为2/10。第二次击球后足球向前偏离，机器人未移脚跟随，第三次颠球在视频约7.36 s失接，随后球触地；最终回合log step 124的服务结果与视频一致。无状态日志，毫秒接触、精确阈值及控制子类无法独立确定。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频帧显示两次击球后球上升、第二次后前飘落地，与stages=2、第3次断点及动态时机/控制归因一致；未发现需改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:55:13Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--nkero7w/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--nkero7w/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "c53c105a52a7786421b9084ab44a059f8bc41526fa6d00e59e5ad1050cad642d",
+   "video_sha256": "c98fc1edcdedc59553058e41a2b3b7424234210fdecd9b0702e2851e780aa4f2",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--nnqazvq.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--nnqazvq.json
new file mode 100644
index 0000000000000000000000000000000000000000..54aee845201794f5481ad589b7870658c4d0359c
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--nnqazvq.json
@@ -0,0 +1,338 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-standard--nnqazvq",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 751,
+   "sim_steps": 2865,
+   "calls": 2818,
+   "rows": 2852,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 3585.1,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.746,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.967,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-standard--nnqazvq/replay.mp4",
+  "duration_s": 7.52,
+  "fps": 25.0,
+  "frames": 188,
+  "sha256": "2352225ccaef0e8a3e336d364213973465be861932aee0cfef324b0ff47d43ef"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–6.92 s，机器人双掌夹住足球，屈膝并将一只脚前伸至球下方，身体保持站立；这些是首次颠球的准备动作，没有已完成的颠球。"
+     },
+     {
+      "t": 7.12,
+      "log_step": 114,
+      "text_zh": "观察：视频 7.12–7.44 s，机器人张开双掌后足球离手下降，来到前伸脚上方；log step 114 的机器人控制调用为 release .16，与这段放球动作相符。"
+     },
+     {
+      "t": 7.48,
+      "log_step": 127,
+      "text_zh": "观察：视频 7.32–7.48 s，足球从双手下方落到前伸腿的小腿/踝轮廓旁，末帧较视频 7.44 s 略高，但未能分辨脚球接触及明确的离脚上升过程。log step 127 的基准服务裁判返回 episode=6、done=true、stages_reached=0、stages_total=10，终止原因为“the ball touched a part other than a foot”；这是服务对最后一回合的结果，不是 agent 的自述。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "视频 0.00–7.08 s，机器人两只平掌从两侧夹压足球，随后张掌放球；候选接触名“夹压”，单掌没有包围或夹爪闭合，现有接触方式不能准确表达两掌对向挤压。此接触属于首次颠球的放球准备。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "未知",
+      "note_zh": "视频 7.32–7.48 s，足球落至机器人前伸腿的小腿/踝轮廓旁，末帧位置略有升高，但远景帧不能确定脚部是否击打足球及具体接触方式。log step 123–124 记录左踝控制动作，不等于已确认脚球接触；log step 127 的服务裁判确认发生非脚部触球，不能据此反推出此前脚部已经击打。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 3.60–7.48 s，机器人抬起并前伸左脚时，右脚持续承重接触草地，直至末帧仍站立。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 127,
+      "text_zh": "观察：视频 0.00–7.48 s 仅见机器人持球准备、放球及球落至前伸腿旁，没有后续连续颠球；log step 127 的最后一回合服务裁判记录 stages_reached=0/10。推断：累计第2次有效颠球未达到；此前置计数也未满足，本次未进入该步骤。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 127,
+      "text_zh": "观察：视频 0.00–7.48 s 仅见机器人持球准备、放球及球落至前伸腿旁，没有后续连续颠球；log step 127 的最后一回合服务裁判记录 stages_reached=0/10。推断：累计第3次有效颠球未达到；此前置计数也未满足，本次未进入该步骤。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 127,
+      "text_zh": "观察：视频 0.00–7.48 s 仅见机器人持球准备、放球及球落至前伸腿旁，没有后续连续颠球；log step 127 的最后一回合服务裁判记录 stages_reached=0/10。推断：累计第4次有效颠球未达到；此前置计数也未满足，本次未进入该步骤。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 127,
+      "text_zh": "观察：视频 0.00–7.48 s 仅见机器人持球准备、放球及球落至前伸腿旁，没有后续连续颠球；log step 127 的最后一回合服务裁判记录 stages_reached=0/10。推断：累计第5次有效颠球未达到；此前置计数也未满足，本次未进入该步骤。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 127,
+      "text_zh": "观察：视频 0.00–7.48 s 仅见机器人持球准备、放球及球落至前伸腿旁，没有后续连续颠球；log step 127 的最后一回合服务裁判记录 stages_reached=0/10。推断：累计第6次有效颠球未达到；此前置计数也未满足，本次未进入该步骤。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 127,
+      "text_zh": "观察：视频 0.00–7.48 s 仅见机器人持球准备、放球及球落至前伸腿旁，没有后续连续颠球；log step 127 的最后一回合服务裁判记录 stages_reached=0/10。推断：累计第7次有效颠球未达到；此前置计数也未满足，本次未进入该步骤。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 127,
+      "text_zh": "观察：视频 0.00–7.48 s 仅见机器人持球准备、放球及球落至前伸腿旁，没有后续连续颠球；log step 127 的最后一回合服务裁判记录 stages_reached=0/10。推断：累计第8次有效颠球未达到；此前置计数也未满足，本次未进入该步骤。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 127,
+      "text_zh": "观察：视频 0.00–7.48 s 仅见机器人持球准备、放球及球落至前伸腿旁，没有后续连续颠球；log step 127 的最后一回合服务裁判记录 stages_reached=0/10。推断：累计第9次有效颠球未达到；此前置计数也未满足，本次未进入该步骤。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 7.48,
+      "log_step": 127,
+      "text_zh": "观察：视频 0.00–7.48 s 仅见机器人持球准备、放球及球落至前伸腿旁，没有后续连续颠球；log step 127 的最后一回合服务裁判记录 stages_reached=0/10。推断：累计第10次有效颠球未达到；此前置计数也未满足，本次未进入该步骤。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "value": null,
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better"
+  },
+  {
+   "name": "ball_height_m",
+   "value": null,
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_1",
+   "t": 7.48,
+   "log_step": 127,
+   "where": "操作中",
+   "evidence_zh": "推断：视频 7.12–7.48 s 中机器人放开双掌，足球下降至前伸腿的小腿/踝轮廓旁，机器人仍站立；log step 127 的本回合基准服务裁判以非脚部触球终止并记录 0/10。因此失败步骤仍是正在尝试且尚未完成的第1次颠球，不把非脚部触球约束或 agent 所称的“下一次触球”作为断点；现有证据不能确定非脚碰撞前是否已有脚部击打。"
+  },
+  "unmet": {
+   "juggles_counted_1": false,
+   "juggles_counted_2": false,
+   "juggles_counted_3": false,
+   "juggles_counted_4": false,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "被挡/碰撞",
+    "t": 7.48,
+    "log_step": 127,
+    "evidence_zh": "推断：视频 7.32–7.48 s 显示足球放落至机器人前伸腿的小腿/踝轮廓旁，未见一次达到要求的离脚上升；log step 127 的本回合服务裁判明确报告球触及非脚部部位并结束回合。结合两者，首次颠球被非脚部碰撞打断；不将其描述为已经脚部击打后的反弹，也不据远景帧指定精确碰撞部位。"
+   }
+  ],
+  "root_causes": [],
+  "gap_zh": "没有状态日志；本轮已查看视频全程的粗采样、视频 6.40–7.48 s 的每帧及视频 7.08–7.48 s 的球腿放大图。25 fps 远景不能确定精确接触部位、毫秒接触顺序或离球速度，末帧之后的裁判终止物理子步未单独呈现。log step 123 的脚本输出在仿真 7.42–7.44 s 为 ankv≈0，log step 124 在仿真 7.45–7.48 s 显示踝部转动；这些只说明动作，不能证明非脚触球发生在击球动作之前。按根因判别顺序，未见目标、顺序或跨阶段策略选错的独立证据；球靠近小腿的画面和非脚触球裁判结果也不足以区分放球落点/脚位的接触几何问题与击球时机、运动执行问题，不能确认接触方案是否可行，故根因留空，不另强定规划或抓取和接触设计。log step 125 的圆拟合坐标为 agent 脚本的图像估计，log step 127 关于首次击打的说法为 agent 自述，均不作为已确认脚部击打的依据。inputs/run.json 的 reward 未单列 touches 或 ball_height_m 原生指标，故 native_metrics 保留 null；有效计数由 log step 127 的服务裁判 stages_reached=0/10 确认。此前五次会话尝试不计入本视频节点进度。"
+ },
+ "summary_zh": "观察：本次评分为第六回合，机器人完成持球准备和放球，球落至前伸腿的小腿/踝轮廓旁；log step 127 的服务裁判记录非脚部触球、有效颠球 0/10，十个节点均未完成。推断：断点仍为第1次颠球操作中，非脚部碰撞使尝试终止，但不能确认此前脚部是否击打足球。没有状态日志，远景与25 fps采样不足以确定接触顺序及根因，故根因留空、置信度由7降至6；成功参考视频仅用于辨认有效颠球的明显离脚上升形态。",
+ "conf": 6,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 6,
+    "summary_zh": "断点（第1次颠球、操作中）可接受，但“前伸脚在7.44–7.48 s击打球、球由降转升”和“击球控制失败”的根因没有证据支持：log step 123–124 显示踝关节在7.44 s前角速度为0、7.45 s才开始转动，而视频7.40–7.48 s球已落在小腿/踝旁，更像是放球后球直接落到非脚部（或脚尚未击球）。",
+    "n_issues": 2,
+    "n_major": 2
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标注与视频和服务裁判结果相符：十个节点均未完成，断点为第1次颠球，球放落到前伸腿旁后被裁判判非脚触球；接触为未知、根因留空并有 gap_zh，均按证据保守处理，未发现需修改之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:03:43Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--nnqazvq/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--nnqazvq/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--nnqazvq/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--nnqazvq/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "d517098725f2383c6c44b62a99efea3dbcb7538c08cdd4fe6a4e2564b3eb194d",
+   "video_sha256": "2352225ccaef0e8a3e336d364213973465be861932aee0cfef324b0ff47d43ef",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--nqgqkrw.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--nqgqkrw.json
new file mode 100644
index 0000000000000000000000000000000000000000..877a8b4b9d81ba948521eb98f95d69c836d9707a
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--nqgqkrw.json
@@ -0,0 +1,308 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-standard--nqgqkrw",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 85,
+   "sim_steps": 256,
+   "calls": 33,
+   "rows": 256,
+   "resets": 2,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 1576.8,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.569,
+   "peak_processes": 10,
+   "peak_disk_gb": 1.178,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-standard--nqgqkrw/replay.mp4",
+  "duration_s": 0.88,
+  "fps": 25.0,
+  "frames": 22,
+  "sha256": "004bf9af8e90ca100f357bb46b387fc1dfdbcfb14f5e891107b85f2f61f273ac"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 0.00–0.84 s，足球先在机器人双掌间，约视频 0.36 s 离掌向前飞出，随后落向前方草地；双脚始终在球后方，没有脚部触球后向上反弹的过程。因此第1次有效颠球未完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "观察：视频 0.00–0.32 s，机器人两侧平掌共同挤压、接触胸前足球，随后球离掌；候选接触名“对向挤压夹持”，单掌没有包围足球，故不能标抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–0.84 s，机器人双脚接触草地承重，末段身体前倾但仍由脚支撑；未见任一脚与足球接触。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 0.00–0.84 s，足球先在机器人双掌间，约视频 0.36 s 离掌向前飞出，随后落向前方草地；双脚始终在球后方，没有脚部触球后向上反弹的过程。因此第2次有效颠球未完成。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 0.00–0.84 s，足球先在机器人双掌间，约视频 0.36 s 离掌向前飞出，随后落向前方草地；双脚始终在球后方，没有脚部触球后向上反弹的过程。因此第3次有效颠球未完成。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 0.00–0.84 s，足球先在机器人双掌间，约视频 0.36 s 离掌向前飞出，随后落向前方草地；双脚始终在球后方，没有脚部触球后向上反弹的过程。因此第4次有效颠球未完成。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 0.00–0.84 s，足球先在机器人双掌间，约视频 0.36 s 离掌向前飞出，随后落向前方草地；双脚始终在球后方，没有脚部触球后向上反弹的过程。因此第5次有效颠球未完成。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 0.00–0.84 s，足球先在机器人双掌间，约视频 0.36 s 离掌向前飞出，随后落向前方草地；双脚始终在球后方，没有脚部触球后向上反弹的过程。因此第6次有效颠球未完成。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 0.00–0.84 s，足球先在机器人双掌间，约视频 0.36 s 离掌向前飞出，随后落向前方草地；双脚始终在球后方，没有脚部触球后向上反弹的过程。因此第7次有效颠球未完成。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 0.00–0.84 s，足球先在机器人双掌间，约视频 0.36 s 离掌向前飞出，随后落向前方草地；双脚始终在球后方，没有脚部触球后向上反弹的过程。因此第8次有效颠球未完成。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 0.00–0.84 s，足球先在机器人双掌间，约视频 0.36 s 离掌向前飞出，随后落向前方草地；双脚始终在球后方，没有脚部触球后向上反弹的过程。因此第9次有效颠球未完成。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 0.00–0.84 s，足球先在机器人双掌间，约视频 0.36 s 离掌向前飞出，随后落向前方草地；双脚始终在球后方，没有脚部触球后向上反弹的过程。因此第10次有效颠球未完成。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "value": null,
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better"
+  },
+  {
+   "name": "ball_height_m",
+   "value": null,
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_1",
+   "t": 0.36,
+   "log_step": null,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 0.32–0.40 s，机器人双掌前方的足球与手分离并继续向前运动；视频 0.40–0.84 s，球远离双脚并下降，末帧落到前方草地附近，机器人没有完成首次脚部击球。推断：首个颠球步骤在放球及衔接脚部击球的过程中失败，断点以约视频 0.36 s 的离手时刻为导航锚点。"
+  },
+  "unmet": {
+   "juggles_counted_1": false,
+   "juggles_counted_2": false,
+   "juggles_counted_3": false,
+   "juggles_counted_4": false,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  },
+  "phenomena": [
+   {
+    "cat": "接触",
+    "item": "没建立接触",
+    "t": 0.64,
+    "log_step": null,
+    "evidence_zh": "观察：视频 0.36–0.84 s，足球从机器人手前向前飞行、下落，机器人双脚留在球后方，球从未与脚接触，也未出现脚击后的上升；所需的首次脚球接触没有建立。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "source_zh": "日志",
+    "evidence_zh": "推断：agent 的关节指令装配存在索引错位。log step 10 的服务 stand 布局前17列是腰部及双臂关节，末2列是无效的 grip；log step 8 给出了对应 start_command。log step 151 的实时脚本却用 start_command.row[2:19] 填入 NOM[12:]，丢掉前2个腰部值并把 grip 值写入腕部目标；log step 155 的最后重试只修改伸腿幅度和循环数，保留该错位。这个目标并非原始持球姿态，与视频 0.00–0.40 s 的手臂运动和球提前向前脱离相符，导致尚未衔接脚部击球便失去可用落球位置；缺少关节跟踪日志，控制子类不再细分。"
+   }
+  ],
+  "gap_zh": "观察：inputs/ 中没有状态日志，评分 reward 也未报告 touches 或 ball_height_m，故两个原生指标保留 null；视频可直接判定没有脚部击球。agent 在 log step 156 自述“伸腿阶段跌倒”，但评分视频 0.00–0.84 s 仅显示球向前离掌、下落及机器人前倾，未显示机器人倒地；以视频为准，不把会话中的跌倒描述标为本次现象。没有逐帧裁判原因或关节跟踪数据，无法进一步分离指令错位与平衡反馈各自的影响。"
+ },
+ "summary_zh": "评分视频中机器人双掌持球后，足球约在视频 0.36 s 向前离掌，随后落向草地，未建立首次脚部触球，十个颠球节点均未完成。首个颠球步骤断在放球与脚部击球的衔接，日志中上半身关节目标索引错位支持控制问题的归因。没有状态日志或原生计数读数，且 agent 的跌倒自述与评分画面不符，故不标倒地现象。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频 0.00–0.84 s 逐帧复核：球约 0.36 s 离掌向前落地，双脚始终未触球，十个节点均未达成，断点、现象与接触标注与画面一致；根因为日志推断（row[2:19] 错位已在 log step 151 核实），已标推断并在 gap_zh 说明限制。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:59:05Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--nqgqkrw/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--nqgqkrw/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "21a65c397bbd9e5c2d5d080b8861a34c0ac1036880f016b678e428893b30e099",
+   "video_sha256": "004bf9af8e90ca100f357bb46b387fc1dfdbcfb14f5e891107b85f2f61f273ac",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--qrxawhd.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--qrxawhd.json
new file mode 100644
index 0000000000000000000000000000000000000000..4a88a2aef079e68b8651aa9ab3d1acb5cf62b4db
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--qrxawhd.json
@@ -0,0 +1,306 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-standard--qrxawhd",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 45,
+   "sim_steps": 6131,
+   "calls": 406,
+   "rows": 4936,
+   "resets": 50,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 4867.8,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-standard--qrxawhd/replay.mp4",
+  "duration_s": 0.48,
+  "fps": 25.0,
+  "frames": 12,
+  "sha256": "48bfd5e1b0ac4efd8fac2c2c06c6ca5728b8a02c5d4198cf01137ea9a5263b3f"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.28,
+      "log_step": 250,
+      "text_zh": "观察：视频 0.00–0.44 s 的全部12帧中，机器人从双掌间放开足球并抬起右腿，球持续下降至草地处，未见离脚向上反弹；视频 0.28、0.32、0.36、0.40、0.44 s 均没有满足首次颠球的升高过程。log step 250 启动最后一次重置后的 kneeflick 控制，log step 251–252 中 agent 自述最后回合球触地、计数为0，与视频相符；不将该自述当作独立裁判计数。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "观察：视频 0.00–0.20 s，双掌在足球两侧夹持后分开释放；候选接触名“对掌挤压夹持”，两只平掌均未包围球，故不是抓握，现有方式无对应项。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–0.44 s，双脚起初着地，右腿抬起后左脚继续接触草地承重；右脚尝试迎向下落足球，但未见球被击起，不据此虚记击打接触。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.44,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.44 s 全部12帧中，足球从双掌间下落至草地处，没有任何一次可见的离脚上升。推断：首次有效颠球未完成，依计数顺序第2次颠球也未达到；视频中没有进入该次数的独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.44,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.44 s 全部12帧中，足球从双掌间下落至草地处，没有任何一次可见的离脚上升。推断：首次有效颠球未完成，依计数顺序第3次颠球也未达到；视频中没有进入该次数的独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.44,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.44 s 全部12帧中，足球从双掌间下落至草地处，没有任何一次可见的离脚上升。推断：首次有效颠球未完成，依计数顺序第4次颠球也未达到；视频中没有进入该次数的独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.44,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.44 s 全部12帧中，足球从双掌间下落至草地处，没有任何一次可见的离脚上升。推断：首次有效颠球未完成，依计数顺序第5次颠球也未达到；视频中没有进入该次数的独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.44,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.44 s 全部12帧中，足球从双掌间下落至草地处，没有任何一次可见的离脚上升。推断：首次有效颠球未完成，依计数顺序第6次颠球也未达到；视频中没有进入该次数的独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.44,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.44 s 全部12帧中，足球从双掌间下落至草地处，没有任何一次可见的离脚上升。推断：首次有效颠球未完成，依计数顺序第7次颠球也未达到；视频中没有进入该次数的独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.44,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.44 s 全部12帧中，足球从双掌间下落至草地处，没有任何一次可见的离脚上升。推断：首次有效颠球未完成，依计数顺序第8次颠球也未达到；视频中没有进入该次数的独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.44,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.44 s 全部12帧中，足球从双掌间下落至草地处，没有任何一次可见的离脚上升。推断：首次有效颠球未完成，依计数顺序第9次颠球也未达到；视频中没有进入该次数的独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.44,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.44 s 全部12帧中，足球从双掌间下落至草地处，没有任何一次可见的离脚上升。推断：首次有效颠球未完成，依计数顺序第10次颠球也未达到；视频中没有进入该次数的独立尝试。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_1",
+   "t": 0.36,
+   "log_step": 250,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 0.08–0.28 s 机器人抬右腿、打开双臂；视频 0.32–0.44 s 足球从腿前继续下落至草地处，未被脚击起，首次颠球没有完成或后续修复。log step 250 是最后一次重置并调用 kneeflick 的动作；log step 251–252 中 agent 报告最后回合球触地且无计数，与视频终态相符。"
+  },
+  "unmet": {
+   "juggles_counted_1": false,
+   "juggles_counted_2": false,
+   "juggles_counted_3": false,
+   "juggles_counted_4": false,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 0.36,
+    "log_step": 250,
+    "evidence_zh": "观察：视频 0.08–0.28 s 机器人已抬起右腿准备迎球，而视频 0.32–0.44 s 足球继续下降，最终到草地处，始终未出现受击后上升。推断：脚的动作未在球下落经过时完成有效拦截，这是对运动目标的迎击失配。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "source_zh": "视频 / 日志",
+    "evidence_zh": "观察：log step 250 中 agent 针对先前脚落在球后下方的问题，自述要提前扫腿并增大髋部前伸，随后运行最后回合 kneeflick；视频 0.08–0.44 s 可见机器人实施抬腿及放球，但足球仍从腿前落至草地，没有有效向上反弹。推断：首次迎球动作的执行控制未能使脚与落球有效配合；现有资料不能分清时序、接触几何与关节跟踪，故不细分控制子类。"
+   }
+  ],
+  "gap_zh": "没有状态日志；inputs/run.json 中 log step 250–251 的服务输出正文为空，末回合落地与0次计数仅有 agent 自述可对照视频。已检查视频全部12帧及末帧，但25 fps单一场边视角不能排除帧间极短的脚球接触，也不能独立核验毫秒接触窗口和速度门槛；未将不确定接触标为实际击打。评分器 reward 未直接报告 touches 或 ball_height_m，两个原生指标值均保留为null；没有证据支持更具体的控制根因。"
+ },
+ "summary_zh": "评分回合只尝试了第1次颠球：机器人从双掌间放球并抬右腿，足球持续下降，视频 0.44 s 已到草地处，十个计数节点均未达到。断点为首次颠球操作中的动态迎击失配，根因保留为控制；多次重置的会话测试不计入本段进度。无状态日志，最后服务输出正文为空，无法进一步判断短暂脚球接触及控制子类。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频仅12帧（0.48 s），球被放开后一路落到草地，未见任何脚击起；10个计数节点均为未达成、断点在第1次颠球、根因为控制，均有帧和log step 250–252支持，未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:00:18Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--qrxawhd/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--qrxawhd/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "834a6be3401ef87906f5a51680b7af4fc4f8fdda804ff944fa79529627b8cff5",
+   "video_sha256": "48bfd5e1b0ac4efd8fac2c2c06c6ca5728b8a02c5d4198cf01137ea9a5263b3f",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--z8zqleu.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--z8zqleu.json
new file mode 100644
index 0000000000000000000000000000000000000000..c0be7109df878b1b4a8df36f4c7f82c4a1515f06
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--z8zqleu.json
@@ -0,0 +1,313 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-standard--z8zqleu",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 100,
+   "sim_steps": 343,
+   "calls": 21,
+   "rows": 343,
+   "resets": 1,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 5026.0,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.096,
+   "peak_processes": 13,
+   "peak_disk_gb": 2.541,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-standard--z8zqleu/replay.mp4",
+  "duration_s": 1.04,
+  "fps": 25.0,
+  "frames": 26,
+  "sha256": "089cdd8c732106cccb0a0dee068d43605eb6e88f0a2c17e981180a05991503aa"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.72,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频0.00–1.00 s（全部26帧）：足球先处在机器人两掌间，视频约0.68–0.80 s双臂打开、右腿抬起；随后足球一直下落，视频1.00 s已在草地处，没有出现脚部击打后离脚上升的完整过程。"
+     },
+     {
+      "t": 1.0,
+      "log_step": 203,
+      "text_zh": "观察：log step 203是最后一次实时服务回合，服务返回“the ball touched the ground”和stages_reached=0/10；与视频末尾足球落到草地、没有有效颠球的画面一致。日志中nkicks=1是agent脚本安排击球的内部次数，不是裁判有效计数。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "观察：视频0.00–约0.64 s机器人双掌从两侧挤住足球，随后张开释放，属于首次颠球的放球准备。候选接触名“对掌挤压夹持”；每只平掌均未包围球，不属于抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.00–约0.68 s双脚承重接触草地；视频约0.72–0.88 s右腿抬起，左脚继续支撑，随后机器人后仰下坠。未凭脚与球在画面上重叠就确认发生击打。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": 203,
+      "text_zh": "观察：视频0.00–1.00 s只出现首次放球及抬腿，随后足球下落到草地，未出现连续颠球；log step 203的最后实时回合服务结果为stages_reached=0/10。推断：按DAG计数顺序，第1次尚未完成，第2次颠球也未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": 203,
+      "text_zh": "观察：视频0.00–1.00 s只出现首次放球及抬腿，随后足球下落到草地，未出现连续颠球；log step 203的最后实时回合服务结果为stages_reached=0/10。推断：按DAG计数顺序，第1次尚未完成，第3次颠球也未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": 203,
+      "text_zh": "观察：视频0.00–1.00 s只出现首次放球及抬腿，随后足球下落到草地，未出现连续颠球；log step 203的最后实时回合服务结果为stages_reached=0/10。推断：按DAG计数顺序，第1次尚未完成，第4次颠球也未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": 203,
+      "text_zh": "观察：视频0.00–1.00 s只出现首次放球及抬腿，随后足球下落到草地，未出现连续颠球；log step 203的最后实时回合服务结果为stages_reached=0/10。推断：按DAG计数顺序，第1次尚未完成，第5次颠球也未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": 203,
+      "text_zh": "观察：视频0.00–1.00 s只出现首次放球及抬腿，随后足球下落到草地，未出现连续颠球；log step 203的最后实时回合服务结果为stages_reached=0/10。推断：按DAG计数顺序，第1次尚未完成，第6次颠球也未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": 203,
+      "text_zh": "观察：视频0.00–1.00 s只出现首次放球及抬腿，随后足球下落到草地，未出现连续颠球；log step 203的最后实时回合服务结果为stages_reached=0/10。推断：按DAG计数顺序，第1次尚未完成，第7次颠球也未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": 203,
+      "text_zh": "观察：视频0.00–1.00 s只出现首次放球及抬腿，随后足球下落到草地，未出现连续颠球；log step 203的最后实时回合服务结果为stages_reached=0/10。推断：按DAG计数顺序，第1次尚未完成，第8次颠球也未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": 203,
+      "text_zh": "观察：视频0.00–1.00 s只出现首次放球及抬腿，随后足球下落到草地，未出现连续颠球；log step 203的最后实时回合服务结果为stages_reached=0/10。推断：按DAG计数顺序，第1次尚未完成，第9次颠球也未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": 203,
+      "text_zh": "观察：视频0.00–1.00 s只出现首次放球及抬腿，随后足球下落到草地，未出现连续颠球；log step 203的最后实时回合服务结果为stages_reached=0/10。推断：按DAG计数顺序，第1次尚未完成，第10次颠球也未达到，且未开始独立尝试。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_1",
+   "t": 0.88,
+   "log_step": 203,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约0.68 s机器人已开始放球、抬右腿尝试首次颠球；视频0.88–1.00 s躯干明显后仰、骨盆快速下降，足球没有被向上颠起而落至草地。log step 203的最后实时服务回合以球触地、0/10结束，之后日志仅有本地仿真测试及代码修改，没有再次运行实时回合来修复第1次颠球。"
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 0.88,
+    "log_step": 203,
+    "evidence_zh": "观察：视频0.80–0.88 s机器人抬腿时身体开始后仰；视频0.92–1.00 s躯干继续倾倒、骨盆下坠、双臂张开，末帧呈接近坐倒姿态，没有恢复站立支撑。足球同时继续下落；最后实时服务结果在log step 203报告球触地。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：log step 194写入、log step 203实际运行的控制器在放球时从站立模式切到lowcmd关节控制，同时安排支撑腿下蹲和右脚逆解摆动；视频约0.72–1.00 s显示这一切换后的抬腿动作未能维持机器人支撑和平衡，首次颠球即失败。可将失稳归于本次全身动作控制，但现有证据不足以进一步区分逆解、跟踪或平衡反馈的具体问题。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "没有状态日志；固定场边视角中，视频约0.80–0.96 s足球与脚部局部重叠，不能确定是否曾发生瞬时脚球接触，也无法核验毫秒级接触历时及速度阈值，因此没有把典型脚部击打写成实际接触。有效颠球为零由全段视频没有离脚上升及log step 203服务0/10共同支持。脚本自身累计的仿真时间与视频时长没有已验证的逐帧映射，所有导航时间均直接取视频；后续本地仿真不是受评回合，未用于补计进度。评分reward没有报告touches或ball_height_m字段，两个原生指标值均保留null。",
+  "unmet": {
+   "juggles_counted_1": false,
+   "juggles_counted_2": false,
+   "juggles_counted_3": false,
+   "juggles_counted_4": false,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  }
+ },
+ "summary_zh": "受评回合只开始了第1次颠球的放球和抬腿，十个计数节点均未完成；视频全段没有足球被有效向上颠起，log step 203的最后实时服务结果为球触地、0/10。机器人在视频约0.88 s抬腿过程中失去平衡并向后下坠，根因可归于全身动作控制未能稳定支撑。没有状态日志，瞬时脚球接触及控制失稳的具体子原因不能确定，后续本地仿真修改未改变受评结果。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标签与视频、log一致：视频全段26帧只有放球、抬腿、后仰坠落，足球落地，无有效颠球，十个节点均未达成；断点在第1次颠球、平衡现象与控制根因均有视频和log step 203支持。未发现需修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:01:31Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--z8zqleu/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--z8zqleu/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "a46320cccb3f2154d5ae0268b05adf52b74d7944dce802a017c1ae210e27430a",
+   "video_sha256": "089cdd8c732106cccb0a0dee068d43605eb6e88f0a2c17e981180a05991503aa",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--zw8ajbm.json b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--zw8ajbm.json
new file mode 100644
index 0000000000000000000000000000000000000000..f3a3a95d7913ae5262cafb8d490ed62a20e40de4
--- /dev/null
+++ b/phase2/embodied-olympics-fun/juggling_feet/trajectories/embodied-olympics-fun-juggling-feet-i00-standard--zw8ajbm.json
@@ -0,0 +1,311 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-juggling-feet-i00-standard--zw8ajbm",
+ "taskKey": "embodied-olympics-fun/juggling_feet",
+ "dag_sha256": "6fb5f868dd0ec1e7ac4f8d9035cd8456789adc45e6c5ceda5fec1ad9d3ba10f6",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 40,
+   "sim_steps": 541,
+   "calls": 52,
+   "rows": 491,
+   "resets": 8,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 5379.8,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.202,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.393,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-juggling-feet-i00-standard--zw8ajbm/replay.mp4",
+  "duration_s": 0.44,
+  "fps": 25.0,
+  "frames": 11,
+  "sha256": "89ab5bbbc7a8a84311516a45b2a7f26dd9c2853defc64daeb7b9ea91b6e54adb"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 10,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "juggles_counted_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": 841,
+      "text_zh": "观察：逐帧检查视频 0.00–0.40 s 的全部11帧，机器人双掌张开放球，右脚前摆，足球持续下落；视频 0.32、0.36、0.40 s 未见足球离脚上升，末帧足球已到草地。log step 841 的最后实时回合 episode 9 返回 outcome=\"the ball touched the ground\"、stages=0/10，与录像一致，第1次有效颠球未完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "双臂协作",
+      "how": "其他",
+      "note_zh": "观察：视频 0.00 s 足球位于两掌之间，随后机器人双掌向两侧分开，视频 0.16–0.20 s 球已落到掌下。候选名称“对掌挤压后放球”：无手指的平掌相向夹持，不是一只手合拢包围物体，故不用抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–0.40 s 机器人双脚起初着地，随后右脚前摆，左脚持续承重接触草地；机器人未跌倒。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 0.28–0.40 s 机器人右脚朝下落足球前摆，尝试“击打”；现有方式无击打项。末帧球与右脚投影贴近，未见球被踢起；25 fps录像不能确认是否有极短的脚球擦碰，不把摆脚记成已完成击球。"
+     }
+    ]
+   },
+   "juggles_counted_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": 841,
+      "text_zh": "观察：视频 0.00–0.40 s 只见机器人首次放球及右脚迎球，足球一路下落至草地，没有连续颠球序列；log step 841 的最后实时回合裁判反馈 stages=0/10。推断：累计有效次数为0，故第2次颠球未达到。"
+     }
+    ]
+   },
+   "juggles_counted_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": 841,
+      "text_zh": "观察：视频 0.00–0.40 s 只见机器人首次放球及右脚迎球，足球一路下落至草地，没有连续颠球序列；log step 841 的最后实时回合裁判反馈 stages=0/10。推断：累计有效次数为0，故第3次颠球未达到。"
+     }
+    ]
+   },
+   "juggles_counted_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": 841,
+      "text_zh": "观察：视频 0.00–0.40 s 只见机器人首次放球及右脚迎球，足球一路下落至草地，没有连续颠球序列；log step 841 的最后实时回合裁判反馈 stages=0/10。推断：累计有效次数为0，故第4次颠球未达到。"
+     }
+    ]
+   },
+   "juggles_counted_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": 841,
+      "text_zh": "观察：视频 0.00–0.40 s 只见机器人首次放球及右脚迎球，足球一路下落至草地，没有连续颠球序列；log step 841 的最后实时回合裁判反馈 stages=0/10。推断：累计有效次数为0，故第5次颠球未达到。"
+     }
+    ]
+   },
+   "juggles_counted_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": 841,
+      "text_zh": "观察：视频 0.00–0.40 s 只见机器人首次放球及右脚迎球，足球一路下落至草地，没有连续颠球序列；log step 841 的最后实时回合裁判反馈 stages=0/10。推断：累计有效次数为0，故第6次颠球未达到。"
+     }
+    ]
+   },
+   "juggles_counted_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": 841,
+      "text_zh": "观察：视频 0.00–0.40 s 只见机器人首次放球及右脚迎球，足球一路下落至草地，没有连续颠球序列；log step 841 的最后实时回合裁判反馈 stages=0/10。推断：累计有效次数为0，故第7次颠球未达到。"
+     }
+    ]
+   },
+   "juggles_counted_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": 841,
+      "text_zh": "观察：视频 0.00–0.40 s 只见机器人首次放球及右脚迎球，足球一路下落至草地，没有连续颠球序列；log step 841 的最后实时回合裁判反馈 stages=0/10。推断：累计有效次数为0，故第8次颠球未达到。"
+     }
+    ]
+   },
+   "juggles_counted_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": 841,
+      "text_zh": "观察：视频 0.00–0.40 s 只见机器人首次放球及右脚迎球，足球一路下落至草地，没有连续颠球序列；log step 841 的最后实时回合裁判反馈 stages=0/10。推断：累计有效次数为0，故第9次颠球未达到。"
+     }
+    ]
+   },
+   "juggles_counted_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": 841,
+      "text_zh": "观察：视频 0.00–0.40 s 只见机器人首次放球及右脚迎球，足球一路下落至草地，没有连续颠球序列；log step 841 的最后实时回合裁判反馈 stages=0/10。推断：累计有效次数为0，故第10次颠球未达到。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "touches",
+   "unit": "count",
+   "threshold": 10,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": "higher_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "juggles_counted_1",
+   "t": 0.36,
+   "log_step": 841,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 0.28–0.40 s 机器人正在摆右脚迎接下落足球，足球仍下降到草地，没有完成首次向上颠球。log step 841 的 episode 9 是最后一次 reset/submit，服务返回触地失败、stages=0/10；log step 943 仍为该回合 done=True、control_steps=40，与 outcome.raw.sim_steps=541 及最后回合前累计501步相符。此处断点是正在尝试的第1次颠球。"
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 0.36,
+    "log_step": 841,
+    "evidence_zh": "观察：视频 0.24–0.40 s 足球从腿旁持续下降，机器人右脚到视频 0.36–0.40 s 才前摆到低处球旁，未在球触地前把球向上送出；log step 841 的服务裁判以球触地终止回合。推断：右脚迎球动作未与来球下落配合，错过有效向上击球窗口。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "观察：log step 841 中 agent 的脚本从起始动作就张开双臂放球，按固定时间插值准备右腿、摆腿及收腿，以 dt=0.01 一次提交动作，执行中没有按球的位置修正时机；服务随即返回触地失败。视频 0.28–0.40 s 显示右脚前摆时足球已下降到脚旁直至草地。推断：本次放球与迎球的运动控制未实现有效向上击打；尚不能把问题进一步唯一归为接触几何、轨迹跟踪或某个参数。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "无模拟器状态日志；已检查全部11帧及成功参考的场边视角。25 fps且球脚在末端投影重叠，无法解析短于一帧的脚球接触、接触力、离球速度和精确错失时刻，故不判定完全未接触，也不细分控制根因。log step 860、912 等是 agent 自建局部模拟器的会话诊断，不是本次评分状态日志。run.json 的 reward 未报告 touches、ball_height_m 两个原生指标，故其值留 null；节点计数依据最后回合服务的 stages=0/10，并由录像中没有上弹核对。",
+  "unmet": {
+   "juggles_counted_1": false,
+   "juggles_counted_2": false,
+   "juggles_counted_3": false,
+   "juggles_counted_4": false,
+   "juggles_counted_5": false,
+   "juggles_counted_6": false,
+   "juggles_counted_7": false,
+   "juggles_counted_8": false,
+   "juggles_counted_9": false,
+   "juggles_counted_10": false
+  }
+ },
+ "summary_zh": "评分回合中机器人双掌放球并摆右脚，但视频 0.00–0.40 s 足球持续下落至草地，log step 841 裁判返回 stages=0/10，因此十个颠球节点均未达到。断点是第1次颠球的操作中：右脚未赶上有效迎球窗口，现象为动态时机，根因保守归为控制。无状态日志且录像仅25 fps，末端是否有极短脚球擦碰及更具体的控制失配机制无法确定。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "录像仅0.44 s（11帧），足球从两掌间放下后一路落到草地，右脚前摆未把球送起；十个节点均未达到，与服务返回 stages=0/10 一致。断点在第1次颠球、操作中，现象“动态时机”与保守的“控制”根因有帧和日志支持，未发现需改动的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:57:14Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--zw8ajbm/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-juggling-feet-i00-standard--zw8ajbm/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "a8e0a12868d71acd545f291f1fe71da1d5418cf3fd52eedb854525b60f457bd9",
+   "video_sha256": "89ab5bbbc7a8a84311516a45b2a7f26dd9c2853defc64daeb7b9ea91b6e54adb",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--jvvbbjw.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--jvvbbjw.json
new file mode 100644
index 0000000000000000000000000000000000000000..e699fa81c1591cf832d002594167d708e42a14fb
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--jvvbbjw.json
@@ -0,0 +1,825 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-privileged--jvvbbjw",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 1317,
+   "sim_steps": 1317,
+   "horizon": 2000,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 20,
+   "stages_total": 20,
+   "fell": 0,
+   "replay_s": 12.14,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 1.205,
+   "metric_bounces": 20.0,
+   "metric_grip_break_max_s": 0.009,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 173,
+   "agent_resets": 8,
+   "agent_replays": 4,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.284,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.255,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 1,
+   "session_log_from_monitor": 1
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--jvvbbjw/replay.mp4",
+  "duration_s": 13.2,
+  "fps": 25.0,
+  "frames": 330,
+  "sha256": "bcd65c47e67f9ae3a5dba4dc77051d90f5128416e28e00dcdcb87ef4d7af83fd"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 20,
+  "final": 20,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.44,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.44,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第1次球拍反弹位于视频约 2.28 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 2.44 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 2.44,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第1次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第1次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第1次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.96,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.96,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第2次球拍反弹位于视频约 2.80 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 2.96 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 2.96,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第2次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第2次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第2次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.52,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.52,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第3次球拍反弹位于视频约 3.36 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 3.52 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 3.52,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第3次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第3次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第3次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.08,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.08,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第4次球拍反弹位于视频约 3.92 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 4.08 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 4.08,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第4次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第4次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第4次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.64,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.64,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第5次球拍反弹位于视频约 4.48 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 4.64 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 4.64,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第5次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第5次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第5次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.2,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.2,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第6次球拍反弹位于视频约 5.04 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 5.20 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 5.2,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第6次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第6次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第6次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.76,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.76,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第7次球拍反弹位于视频约 5.60 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 5.76 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 5.76,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第7次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第7次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第7次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.32,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.32,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第8次球拍反弹位于视频约 6.16 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 6.32 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 6.32,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第8次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第8次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第8次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.88,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.88,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第9次球拍反弹位于视频约 6.72 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 6.88 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 6.88,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第9次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第9次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第9次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.44,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.44,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第10次球拍反弹位于视频约 7.28 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 7.44 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 7.44,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第10次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第10次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第10次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.0,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.0,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第11次球拍反弹位于视频约 7.88 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 8.00 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 8.0,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第11次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第11次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第11次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.56,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.56,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第12次球拍反弹位于视频约 8.44 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 8.56 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 8.56,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第12次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第12次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第12次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.16,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.16,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第13次球拍反弹位于视频约 9.00 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 9.16 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 9.16,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第13次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第13次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第13次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.76,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.76,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第14次球拍反弹位于视频约 9.60 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 9.76 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 9.76,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第14次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第14次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第14次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.32,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.32,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第15次球拍反弹位于视频约 10.16 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 10.32 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 10.32,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第15次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第15次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第15次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.92,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.92,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第16次球拍反弹位于视频约 10.76 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 10.92 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 10.92,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第16次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第16次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第16次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [
+     [
+      11.48,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 11.48,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第17次球拍反弹位于视频约 11.32 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 11.48 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 11.48,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第17次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第17次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第17次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [
+     [
+      12.04,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 12.04,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第18次球拍反弹位于视频约 11.88 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 12.04 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 12.04,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第18次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第18次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第18次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [
+     [
+      12.6,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 12.6,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第19次球拍反弹位于视频约 12.44 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 12.60 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 12.6,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第19次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第19次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第19次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [
+     [
+      13.16,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 13.16,
+      "log_step": null,
+      "text_zh": "观察：按受评视频中的反弹顺序计数，第20次球拍反弹位于视频约 13.04 s；右手持红色球拍的机器人使下落橙色球在拍面附近转为上升，至视频约 13.16 s 回到开局释放杯的高度附近。该次触拍前后已逐帧检查，节点时间为画面近似值，误差约0.1秒。"
+     },
+     {
+      "t": 13.16,
+      "log_step": null,
+      "text_zh": "推断：inputs/run.json 的受评 reward.metric_bounces=20、stages_reached=20、success=1、replay_success=1、deterministic=1，与视频中20次连续反弹一致，因此第20次有效颠球已被裁判接受并锁存；速率比及毫秒级接触阈值由此判定，未从25 fps画面直接测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：第20次反弹中，机器人右手持红色球拍，以拍面击打橙色小球，球由下降转为上升。候选接触方式为“击打”，现有接触方式没有对应值，故用其他；手握拍柄属于同一工具链，不另重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第20次颠球前后的场地全景中，机器人双脚着地支撑身体，保持站立并用右臂持拍。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": 20.0
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": 1.205
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.009
+  }
+ ],
+ "failure": null,
+ "summary_zh": "受评视频中，机器人先将右手所持红色球拍调平，在视频约 2.00 s 放球后连续完成20次反弹，第1次与第20次有效颠球分别定位于视频约 2.44 s 和 13.16 s，全部20个事件节点最终成立。机器人在所检画面中持续持拍站立，评分器报告 metric_bounces=20、stages_reached=20、success=1 且重放成功、确定性一致，无失败断点。没有状态日志或受评逐次裁判事件表，节点时间由视频逐帧近似定位；0.98速率比及毫秒级接触验收依赖最终裁判结果，本次自身录像不作独立oracle比较。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "成功运行（success=1，stages 20/20，metric_bounces=20）。20个反弹节点全部成立，视频时间与帧一致，无失败断点；接触标注（手持工具·击打→其他、腿/躯干·支撑）合理。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:53:07Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--jvvbbjw/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--jvvbbjw/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "8bcc985f290700661b6ae3503573495dec80faac4764e6ba4526a86979954317",
+   "video_sha256": "bcd65c47e67f9ae3a5dba4dc77051d90f5128416e28e00dcdcb87ef4d7af83fd",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--pjyhf7x.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--pjyhf7x.json
new file mode 100644
index 0000000000000000000000000000000000000000..743da8de2d522e8c536818d1a41fc346ee5a9edc
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--pjyhf7x.json
@@ -0,0 +1,825 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-privileged--pjyhf7x",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 1316,
+   "sim_steps": 1316,
+   "horizon": 2000,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 20,
+   "stages_total": 20,
+   "fell": 0,
+   "replay_s": 10.76,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 1.205,
+   "metric_bounces": 20.0,
+   "metric_grip_break_max_s": 0.015,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 108,
+   "agent_resets": 66,
+   "agent_replays": 5,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.227,
+   "peak_processes": 13,
+   "peak_disk_gb": 1.016,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--pjyhf7x/replay.mp4",
+  "duration_s": 13.2,
+  "fps": 25.0,
+  "frames": 330,
+  "sha256": "3da9549824163b8a7e1edde400791bebbd5f82d7d66c1afef431d3e1bf7c3ecd"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 20,
+  "final": 20,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.32,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.32,
+      "log_step": null,
+      "text_zh": "观察：视频 2.20–2.44 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 2.32,
+      "log_step": 189,
+      "text_zh": "推断：视频 2.20 s 首次触拍后，小球在视频 2.32 s 已升回开局释放杯球心高度附近，并继续上升至视频 2.44 s；结合评分器确认的20次有效颠球及首击后的明显增高，将首次完成定位于约视频 2.32 s。速度比和毫秒接触持续时间由裁判判定，25 fps画面不能独立复核。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.20–2.44 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.32 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.28,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.28,
+      "log_step": null,
+      "text_zh": "观察：视频 3.12–3.36 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 3.28,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第2次有效颠球最终成立。该节点的约视频 3.28 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 3.12–3.36 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 3.28 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.72,
+      "log_step": null,
+      "text_zh": "观察：视频 3.60–3.84 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 3.72,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第3次有效颠球最终成立。该节点的约视频 3.72 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 3.60–3.84 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 3.72 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.24,
+      "log_step": null,
+      "text_zh": "观察：视频 4.08–4.32 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 4.24,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第4次有效颠球最终成立。该节点的约视频 4.24 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 4.08–4.32 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 4.24 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.72,
+      "log_step": null,
+      "text_zh": "观察：视频 4.56–4.80 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 4.72,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第5次有效颠球最终成立。该节点的约视频 4.72 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 4.56–4.80 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 4.72 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.16,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.16,
+      "log_step": null,
+      "text_zh": "观察：视频 5.04–5.28 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 5.16,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第6次有效颠球最终成立。该节点的约视频 5.16 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 5.04–5.28 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 5.16 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.68,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.68,
+      "log_step": null,
+      "text_zh": "观察：视频 5.52–5.80 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 5.68,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第7次有效颠球最终成立。该节点的约视频 5.68 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 5.52–5.80 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 5.68 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.16,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.16,
+      "log_step": null,
+      "text_zh": "观察：视频 6.04–6.28 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 6.16,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第8次有效颠球最终成立。该节点的约视频 6.16 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 6.04–6.28 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 6.16 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.68,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.68,
+      "log_step": null,
+      "text_zh": "观察：视频 6.56–6.80 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 6.68,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第9次有效颠球最终成立。该节点的约视频 6.68 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 6.56–6.80 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 6.68 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.2,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.2,
+      "log_step": null,
+      "text_zh": "观察：视频 7.04–7.32 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 7.2,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第10次有效颠球最终成立。该节点的约视频 7.20 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 7.04–7.32 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 7.20 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.68,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.68,
+      "log_step": null,
+      "text_zh": "观察：视频 7.56–7.80 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 7.68,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第11次有效颠球最终成立。该节点的约视频 7.68 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 7.56–7.80 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 7.68 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.2,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.2,
+      "log_step": null,
+      "text_zh": "观察：视频 8.04–8.32 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 8.2,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第12次有效颠球最终成立。该节点的约视频 8.20 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 8.04–8.32 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 8.20 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.68,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.68,
+      "log_step": null,
+      "text_zh": "观察：视频 8.56–8.80 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 8.68,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第13次有效颠球最终成立。该节点的约视频 8.68 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 8.56–8.80 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 8.68 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.2,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.2,
+      "log_step": null,
+      "text_zh": "观察：视频 9.04–9.32 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 9.2,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第14次有效颠球最终成立。该节点的约视频 9.20 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 9.04–9.32 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 9.20 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.72,
+      "log_step": null,
+      "text_zh": "观察：视频 9.56–9.84 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 9.72,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第15次有效颠球最终成立。该节点的约视频 9.72 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 9.56–9.84 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 9.72 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.2,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.2,
+      "log_step": null,
+      "text_zh": "观察：视频 10.08–10.36 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 10.2,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第16次有效颠球最终成立。该节点的约视频 10.20 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 10.08–10.36 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 10.20 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [
+     [
+      11.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 11.24,
+      "log_step": null,
+      "text_zh": "观察：视频 11.12–11.36 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 11.24,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第17次有效颠球最终成立。该节点的约视频 11.24 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 11.12–11.36 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 11.24 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [
+     [
+      11.76,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 11.76,
+      "log_step": null,
+      "text_zh": "观察：视频 11.64–11.88 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 11.76,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第18次有效颠球最终成立。该节点的约视频 11.76 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 11.64–11.88 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 11.76 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [
+     [
+      12.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 12.24,
+      "log_step": null,
+      "text_zh": "观察：视频 12.12–12.36 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 12.24,
+      "log_step": 189,
+      "text_zh": "推断：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1，log step 189 对最终1316行交付文件的重放也列出 bounce 1 至 bounce 20，故第19次有效颠球最终成立。该节点的约视频 12.24 s 是结合反弹顺序与最终有效总数估计的上升阶段导航时刻；画面无逐次裁判计数，不能逐帧验证速度比，局部计数归属可能前后相差一轮（约视频0.5 s）。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 12.12–12.36 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 12.24 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [
+     [
+      13.16,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 13.16,
+      "log_step": null,
+      "text_zh": "观察：视频 13.08–13.16 s，橙色小球从机器人右手所持红色拍面附近向上离开，升至原释放杯球心高度附近或更高；机器人保持站立，球拍留在右手。"
+     },
+     {
+      "t": 13.16,
+      "log_step": 189,
+      "text_zh": "观察：评分器 reward 的 metric_bounces=20、stages_reached=20、success=1、deterministic=1；log step 189 对最终1316行交付文件的独立重放报告全部20个阶段及 success=True，log step 191 两个新进程的终态哈希相同。视频 13.04 s 小球降至拍面附近，视频 13.08–13.16 s 再次上升并在视频末帧接近原释放高度，故以视频 13.16 s 标记第20次完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 13.08–13.16 s，机器人经右手握持的球拍红色拍面击起橙色小球。候选方式为“击打”；现有方式词表没有击打项，故用其他，持拍柄不重复计入接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 13.16 s 所在反弹过程中，机器人双脚接地、双腿支撑站立，同时右臂持拍颠球。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": 20.0
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": 1.205
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.015
+  }
+ ],
+ "failure": null,
+ "summary_zh": "观察：视频 0–2.04 s 机器人将右手球拍举至近水平并等待放球，此后持续用拍面击起橙色小球，至视频 13.16 s 仍持拍站立；评分器确认20/20次有效颠球、success=1，最终交付文件重放验证一致（log step 189、191）。推断：近水平拍面的小幅抬动及跟随使颠球持续，20个事件节点均锁存完成，无失败断点。观察：没有状态日志或逐次裁判阶段时间，逐帧可见约23次触拍，不能把它们逐次等同于20次有效计数；中间节点时刻结合反弹顺序与总数估计，局部有效计数归属可能相差一轮（约视频0.5 s），0.98速度比及毫秒接触门槛仅由最终裁判结果确认。",
+ "conf": 6,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "成功运行（评分器 success=1，20/20 阶段，无失败）；20个颠球节点均为 final=true，接触标注（其他·击打 + 腿/躯干·支撑）与DAG和视频一致，未见需修改的内容。逐次时刻为估计值，文件已明确标注。校验器通过。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:52:23Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--pjyhf7x/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--pjyhf7x/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "907205e4c34cebdbb25ae3270559bcbf3ae11c35bca24a0a07106aa5a73a093d",
+   "video_sha256": "3da9549824163b8a7e1edde400791bebbd5f82d7d66c1afef431d3e1bf7c3ecd",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--tgekfy7.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--tgekfy7.json
new file mode 100644
index 0000000000000000000000000000000000000000..ff653ecd408b4b377c788f1a103610d80a4bd316
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--tgekfy7.json
@@ -0,0 +1,755 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-privileged--tgekfy7",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.7,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 1172,
+   "sim_steps": 1172,
+   "horizon": 2000,
+   "episode_over": 1,
+   "progress": 0.7,
+   "stages_reached": 14,
+   "stages_total": 20,
+   "fell": 0,
+   "replay_s": 10.44,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 0.004,
+   "metric_bounces": 14.0,
+   "metric_grip_break_max_s": 0.008,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 952,
+   "agent_resets": 27,
+   "agent_replays": 22,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 2.136,
+   "peak_processes": 11,
+   "peak_disk_gb": 0.864,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--tgekfy7/replay.mp4",
+  "duration_s": 11.76,
+  "fps": 25.0,
+  "frames": 294,
+  "sha256": "7f4b813efb977c2587974051e5441b208bf0b2bf6f66c22792c5a7fc70db2752"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 14,
+  "final": 14,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.4,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.4,
+      "log_step": 434,
+      "text_zh": "观察：视频 2.28–2.40 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第1次反弹；视频 2.40 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 1”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 2.4,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.28–2.40 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.28–2.40 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.0,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 434,
+      "text_zh": "观察：视频 2.92–3.00 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第2次反弹；视频 3.00 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 2”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 3.0,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.92–3.00 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.92–3.00 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.6,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.6,
+      "log_step": 434,
+      "text_zh": "观察：视频 3.48–3.60 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第3次反弹；视频 3.60 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 3”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 3.6,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 3.48–3.60 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 3.48–3.60 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.12,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.12,
+      "log_step": 434,
+      "text_zh": "观察：视频 4.04–4.12 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第4次反弹；视频 4.12 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 4”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 4.12,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 4.04–4.12 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 4.04–4.12 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.68,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.68,
+      "log_step": 434,
+      "text_zh": "观察：视频 4.60–4.68 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第5次反弹；视频 4.68 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 5”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 4.68,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 4.60–4.68 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 4.60–4.68 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.2,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.2,
+      "log_step": 434,
+      "text_zh": "观察：视频 5.08–5.20 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第6次反弹；视频 5.20 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 6”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 5.2,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 5.08–5.20 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 5.08–5.20 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.72,
+      "log_step": 434,
+      "text_zh": "观察：视频 5.64–5.72 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第7次反弹；视频 5.72 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 7”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 5.72,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 5.64–5.72 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 5.64–5.72 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.28,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.28,
+      "log_step": 434,
+      "text_zh": "观察：视频 6.24–6.28 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第8次反弹；视频 6.28 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 8”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 6.28,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 6.24–6.28 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 6.24–6.28 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.96,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.96,
+      "log_step": 434,
+      "text_zh": "观察：视频 6.88–6.96 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第9次反弹；视频 6.96 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 9”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 6.96,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 6.88–6.96 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 6.88–6.96 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.68,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.68,
+      "log_step": 434,
+      "text_zh": "观察：视频 7.64–7.68 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第10次反弹；视频 7.68 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 10”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 7.68,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 7.64–7.68 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 7.64–7.68 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.44,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.44,
+      "log_step": 434,
+      "text_zh": "观察：视频 8.36–8.44 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第11次反弹；视频 8.44 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 11”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 8.44,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 8.36–8.44 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 8.36–8.44 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.2,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.2,
+      "log_step": 434,
+      "text_zh": "观察：视频 9.16–9.20 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第12次反弹；视频 9.20 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 12”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 9.2,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 9.16–9.20 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 9.16–9.20 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.0,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.0,
+      "log_step": 434,
+      "text_zh": "观察：视频 9.96–10.00 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第13次反弹；视频 10.00 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 13”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 10.0,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 9.96–10.00 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 9.96–10.00 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.72,
+      "log_step": 434,
+      "text_zh": "观察：视频 10.68–10.72 s，橙色小球在机器人右手所持红色球拍处由下降转为上升，完成顺序中的第14次反弹；视频 10.72 s 已上升至开局释放高度附近或以上。log step 434 最终交付文件的重放裁判列出已锁存的“bounce 14”；评分器 metric_bounces=14、stages_reached=14。"
+     },
+     {
+      "t": 10.72,
+      "log_step": 434,
+      "text_zh": "推断：逐帧反弹顺序与最终交付重放的前14个阶段一致，结合裁判锁存确认该次有效；时间取已检查的上升画面，精度约0.1秒，不用视频估算0.98速率比或毫秒级触球时长。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 10.68–10.72 s，机器人右手握持球拍，用红色拍面击起橙色小球；候选方式“击打”，现有方式词表无对应项，故用其他；不重复标持拍手的握柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 10.68–10.72 s，机器人双足接地支撑身体，右臂完成该次颠球。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 11.44,
+      "log_step": 434,
+      "text_zh": "观察：视频 10.72 s 后机器人继续右手持拍追球；视频 11.20–11.72 s 身体明显向后倾倒，下降的小球从拍面一侧漏过，随后落至地面，没有第15次向上反弹。log step 434 最终交付重放裁判仅锁存 bounce 1–14，结束原因为“the ball touched the floor”，与评分器 metric_bounces=14 一致。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 10.72–11.12 s，机器人追第15次回球时双足仍接地支撑，随后身体持续后倾、足部支撑失稳；右手仍持球拍，但本次没有建立拍面与小球的击打接触，不虚构工具击球接触。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 11.72,
+      "log_step": 434,
+      "text_zh": "推断：最终评分器 metric_bounces=14，log step 434 交付重放裁判只有 bounce 1–14；视频 11.44–11.72 s 小球漏过球拍落地且没有恢复颠球，因此按原生顺序计数，第16次颠球未达到，亦未进入该步。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 11.72,
+      "log_step": 434,
+      "text_zh": "推断：最终评分器 metric_bounces=14，log step 434 交付重放裁判只有 bounce 1–14；视频 11.44–11.72 s 小球漏过球拍落地且没有恢复颠球，因此按原生顺序计数，第17次颠球未达到，亦未进入该步。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 11.72,
+      "log_step": 434,
+      "text_zh": "推断：最终评分器 metric_bounces=14，log step 434 交付重放裁判只有 bounce 1–14；视频 11.44–11.72 s 小球漏过球拍落地且没有恢复颠球，因此按原生顺序计数，第18次颠球未达到，亦未进入该步。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 11.72,
+      "log_step": 434,
+      "text_zh": "推断：最终评分器 metric_bounces=14，log step 434 交付重放裁判只有 bounce 1–14；视频 11.44–11.72 s 小球漏过球拍落地且没有恢复颠球，因此按原生顺序计数，第19次颠球未达到，亦未进入该步。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 11.72,
+      "log_step": 434,
+      "text_zh": "推断：最终评分器 metric_bounces=14，log step 434 交付重放裁判只有 bounce 1–14；视频 11.44–11.72 s 小球漏过球拍落地且没有恢复颠球，因此按原生顺序计数，第20次颠球未达到，亦未进入该步。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": 14.0
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": 0.004
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.008
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_15",
+   "t": 11.2,
+   "log_step": 434,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 10.72 s 完成第14次反弹后，机器人继续持拍迎接下落球；视频 11.20 s 起身体后倾加剧，视频 11.44–11.52 s 小球在拍面一侧继续下落，直至视频 11.72 s 结束均未再击起。log step 434 确认所交付1172行轨迹的裁判只计14次，并因球触地结束；失败步骤是正在尝试的第15次颠球。"
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 11.2,
+    "log_step": null,
+    "evidence_zh": "观察：视频 11.12–11.72 s，机器人持拍追球时躯干持续向后倾斜、骨盆下降，双腿向前伸出，末尾呈向后坐倒的失衡姿态，未恢复站立。此处标可见的支撑失稳；评分器 fell=0，交付重放裁判在球触地时先结束，并未记录跌倒判负。"
+   },
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 11.44,
+    "log_step": 434,
+    "evidence_zh": "观察：视频 11.36–11.60 s，橙色小球下降时，机器人仍在调整右手球拍，但拍面没有跟到球的下降路径，球从旁边漏过并继续落地，没有下一次反弹。log step 434 最终重放裁判计数停在14，报告球触地。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：log step 425、429、432 的轨迹生成代码以预测球在目标高度的落点来修正右臂及腰部关节，并用 stand 动作继续追球；log step 434 将这一延长追球轨迹作为最终交付并重放。视频 10.72–11.72 s 显示该控制序列既未维持身体稳定支撑，也未把拍面接到下一次下落球，故归因控制。现有证据不足以进一步区分逆解、反馈跟踪或接触参数问题。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志；25 fps视频不能独立核实0.02秒触球合并、0.98出入球速率比及0.05秒持拍接触门槛，前14次有效性结合最终交付重放的裁判阶段和评分器总数确认。无法从现有输出进一步确定控制失败的具体子类，也不把会话中其他候选轨迹的碰撞或跌倒混入本次轨迹。",
+  "unmet": {
+   "bounce_1": true,
+   "bounce_2": true,
+   "bounce_3": true,
+   "bounce_4": true,
+   "bounce_5": true,
+   "bounce_6": true,
+   "bounce_7": true,
+   "bounce_8": true,
+   "bounce_9": true,
+   "bounce_10": true,
+   "bounce_11": true,
+   "bounce_12": true,
+   "bounce_13": true,
+   "bounce_14": true,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  }
+ },
+ "summary_zh": "机器人完成并锁存前14次有效颠球，最终为14/20；视频 2.40–10.72 s 的反弹顺序与 log step 434 最终交付重放的14个裁判阶段一致。断点为第15次颠球：视频 11.20–11.72 s 机器人追球时向后失衡，拍面未接上下降球，球触地结束，根因支持到控制层。无逐帧状态日志，毫秒接触与速率验收依赖原生裁判，控制具体子类不能确定。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "前14次颠球时间与接触标注与视频一致（2.2–2.4 s 击起、10.72 s 第14次），断点为第15次、视频 11.2 s 起向后坐倒、球从拍旁漏过并触地，与评分器 metric_bounces=14 一致；平衡+动态时机、控制根因有帧支持。未发现需修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:58:16Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--tgekfy7/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--tgekfy7/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "23fe6c9ed36ed22c5c9c02428f45ad44a14853a4957f353fc8700288b5ad0f84",
+   "video_sha256": "7f4b813efb977c2587974051e5441b208bf0b2bf6f66c22792c5a7fc70db2752",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--wyezdyx.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--wyezdyx.json
new file mode 100644
index 0000000000000000000000000000000000000000..8b7335b997203457913724e90b2e9965b2477050
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--wyezdyx.json
@@ -0,0 +1,823 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-privileged--wyezdyx",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 1202,
+   "sim_steps": 1202,
+   "horizon": 2000,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 20,
+   "stages_total": 20,
+   "fell": 0,
+   "replay_s": 12.69,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 1.207,
+   "metric_bounces": 20.0,
+   "metric_grip_break_max_s": 0.003,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 1595,
+   "agent_resets": 0,
+   "agent_replays": 26,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--wyezdyx/replay.mp4",
+  "duration_s": 12.04,
+  "fps": 25.0,
+  "frames": 301,
+  "sha256": "7ee0687815fe5b721696d80f0ccb1230d95af93c64daaa4e9960bee356694b4f"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 20,
+  "final": 20,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.32,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.32,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 2.16–2.32 s，机器人右手所持红色拍面在视频约 2.20 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第1次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 2.32,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第1次对应为有效计数 bounce 1 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 2.20 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 2.16–2.32 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.84,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 2.64–2.84 s，机器人右手所持红色拍面在视频约 2.68 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第2次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 2.84,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第2次对应为有效计数 bounce 2 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 2.68 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 2.64–2.84 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.32,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.32,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 3.16–3.32 s，机器人右手所持红色拍面在视频约 3.20 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第3次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 3.32,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第3次对应为有效计数 bounce 3 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 3.20 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 3.16–3.32 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.84,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 3.64–3.84 s，机器人右手所持红色拍面在视频约 3.68 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第4次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 3.84,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第4次对应为有效计数 bounce 4 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 3.68 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 3.64–3.84 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.32,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.32,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 4.16–4.32 s，机器人右手所持红色拍面在视频约 4.20 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第5次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 4.32,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第5次对应为有效计数 bounce 5 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 4.20 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 4.16–4.32 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.84,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 4.68–4.84 s，机器人右手所持红色拍面在视频约 4.72 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第6次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 4.84,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第6次对应为有效计数 bounce 6 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 4.72 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 4.68–4.84 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.32,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.32,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 5.20–5.32 s，机器人右手所持红色拍面在视频约 5.24 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第7次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 5.32,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第7次对应为有效计数 bounce 7 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 5.24 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 5.20–5.32 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.84,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 5.68–5.84 s，机器人右手所持红色拍面在视频约 5.72 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第8次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 5.84,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第8次对应为有效计数 bounce 8 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 5.72 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 5.68–5.84 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.32,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.32,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 6.20–6.32 s，机器人右手所持红色拍面在视频约 6.24 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第9次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 6.32,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第9次对应为有效计数 bounce 9 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 6.24 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 6.20–6.32 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.84,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 6.72–6.84 s，机器人右手所持红色拍面在视频约 6.76 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第10次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 6.84,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第10次对应为有效计数 bounce 10 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 6.76 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 6.72–6.84 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.4,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.4,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 7.24–7.40 s，机器人右手所持红色拍面在视频约 7.28 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第11次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 7.4,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第11次对应为有效计数 bounce 11 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 7.28 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 7.24–7.40 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.92,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.92,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 7.76–7.92 s，机器人右手所持红色拍面在视频约 7.80 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第12次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 7.92,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第12次对应为有效计数 bounce 12 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 7.80 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 7.76–7.92 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.44,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.44,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 8.28–8.44 s，机器人右手所持红色拍面在视频约 8.32 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第13次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 8.44,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第13次对应为有效计数 bounce 13 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 8.32 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 8.28–8.44 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.96,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.96,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 8.80–8.96 s，机器人右手所持红色拍面在视频约 8.84 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第14次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 8.96,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第14次对应为有效计数 bounce 14 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 8.84 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 8.80–8.96 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.48,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.48,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 9.32–9.48 s，机器人右手所持红色拍面在视频约 9.36 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第15次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 9.48,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第15次对应为有效计数 bounce 15 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 9.36 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 9.32–9.48 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.0,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.0,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 9.84–10.00 s，机器人右手所持红色拍面在视频约 9.88 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第16次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 10.0,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第16次对应为有效计数 bounce 16 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 9.88 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 9.84–10.00 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.52,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.52,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 10.36–10.52 s，机器人右手所持红色拍面在视频约 10.40 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第17次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 10.52,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第17次对应为有效计数 bounce 17 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 10.40 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 10.36–10.52 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [
+     [
+      11.04,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 11.04,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 10.88–11.04 s，机器人右手所持红色拍面在视频约 10.92 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第18次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 11.04,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第18次对应为有效计数 bounce 18 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 10.92 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 10.88–11.04 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [
+     [
+      11.56,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 11.56,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 11.40–11.56 s，机器人右手所持红色拍面在视频约 11.44 s 附近把下落橙球击起，球随后离拍并升至开局释放杯中球心的高度附近或更高；这是按录像顺序数到的第19次击球上升。机器人双脚保持着地支撑；所给完成锚点精度约0.1 s。"
+     },
+     {
+      "t": 11.56,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第19次对应为有效计数 bounce 19 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 11.44 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 11.40–11.56 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [
+     [
+      12.0,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 12.0,
+      "log_step": null,
+      "text_zh": "观察：视频 11.88–12.00 s，机器人右手持红色球拍完成录像顺序的第20次击球；视频 11.92–11.96 s 球贴近拍面后反向上升，视频 12.00 s 最后一帧球已离拍、仍在上升，机器人保持站立和持拍。画面到此结束，未显示该次完整飞行。"
+     },
+     {
+      "t": 12.0,
+      "log_step": null,
+      "text_zh": "推断：本次交付的评分器记录 success=1、replay_success=1、deterministic=1、metric_bounces=20、stages_reached=20/20；结合逐帧可辨的20次击球上升，将第20次对应为有效计数 bounce 20 并按事件节点锁存。25 fps录像不能独立验证0.98速率比及0.02秒接触门槛，这些验收依评分器确认。第20次以视频 12.00 s 最后一帧作为近似完成锚点；其最终达标由评分结果确认，不能从末帧独立测出精确达标时刻。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "机器人右手持红色球拍，拍面在视频约 11.96 s 附近击打橙色乒乓球并使其向上离拍。候选方式“击打”，现有接触方式词表无此项，依S16用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "视频 11.92–12.00 s 对应击球期间，机器人双脚着地支撑身体，未见迈步或跌倒。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": 20.0
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": 1.207
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.003
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人右手持红色球拍连续颠球，完成全部20个事件节点；评分器给出20次有效计数、成功且重放一致，没有失败断点。已逐帧检查完整录像，视频约 2.20–11.96 s 可辨20次击球，橙球反复向上离拍，机器人始终站立持拍。没有状态日志，25 fps录像无法独立核验毫秒级接触与精确速率；第20次在视频 12.00 s 末帧仍上升，其完成由本次评分器确认，节点时间为近似视频锚点。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "成功运行，评分器 20/20、未摔倒；20个 bounce 节点均达成、无断点，击球时间锚点与录像一致，接触标注（其他+击打、腿/躯干支撑）符合 S16。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:52:20Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--wyezdyx/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--wyezdyx/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "00ef236981a515cdb5701b81f619c0d303c077fd89b00150e64f1b06dfc30a32",
+   "video_sha256": "7ee0687815fe5b721696d80f0ccb1230d95af93c64daaa4e9960bee356694b4f",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--xemhvcr.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--xemhvcr.json
new file mode 100644
index 0000000000000000000000000000000000000000..67614a96960fc1cc1b9adb2ab3b2f97e28664d8f
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--xemhvcr.json
@@ -0,0 +1,830 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-privileged--xemhvcr",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 1395,
+   "sim_steps": 1395,
+   "horizon": 2000,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 20,
+   "stages_total": 20,
+   "fell": 0,
+   "replay_s": 10.94,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 1.202,
+   "metric_bounces": 20.0,
+   "metric_grip_break_max_s": 0.004,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 15,
+   "agent_resets": 0,
+   "agent_replays": 3,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.449,
+   "peak_processes": 8,
+   "peak_disk_gb": 0.241,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--xemhvcr/replay.mp4",
+  "duration_s": 13.96,
+  "fps": 25.0,
+  "frames": 349,
+  "sha256": "65c0d1f540557233bd79b706d3d70fed38d1c2adf76b1a5c45d70ab3b2d5c36c"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 20,
+  "final": 20,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.4,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.4,
+      "log_step": null,
+      "text_zh": "观察：视频 2.24–2.32 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 2.40 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第1次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 2.4,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第1次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.24–2.32 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第1次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.08,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.08,
+      "log_step": null,
+      "text_zh": "观察：视频 2.88–2.96 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 3.08 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第2次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 3.08,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第2次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.88–2.96 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第2次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.68,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": null,
+      "text_zh": "观察：视频 3.48–3.56 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 3.68 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第3次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 3.68,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第3次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 3.48–3.56 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第3次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.28,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.28,
+      "log_step": null,
+      "text_zh": "观察：视频 4.08–4.16 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 4.28 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第4次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 4.28,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第4次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 4.08–4.16 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第4次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.88,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.88,
+      "log_step": null,
+      "text_zh": "观察：视频 4.68–4.76 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 4.88 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第5次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 4.88,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第5次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 4.68–4.76 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第5次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.52,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.52,
+      "log_step": null,
+      "text_zh": "观察：视频 5.32–5.40 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 5.52 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第6次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 5.52,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第6次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 5.32–5.40 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第6次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.12,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.12,
+      "log_step": null,
+      "text_zh": "观察：视频 5.92–6.00 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 6.12 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第7次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 6.12,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第7次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 5.92–6.00 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第7次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.72,
+      "log_step": null,
+      "text_zh": "观察：视频 6.52–6.60 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 6.72 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第8次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 6.72,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第8次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 6.52–6.60 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第8次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.32,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.32,
+      "log_step": null,
+      "text_zh": "观察：视频 7.12–7.20 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 7.32 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第9次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 7.32,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第9次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 7.12–7.20 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第9次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.92,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.92,
+      "log_step": null,
+      "text_zh": "观察：视频 7.72–7.80 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 7.92 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第10次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 7.92,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第10次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 7.72–7.80 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第10次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.52,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.52,
+      "log_step": null,
+      "text_zh": "观察：视频 8.32–8.40 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 8.52 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第11次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 8.52,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第11次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 8.32–8.40 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第11次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.12,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.12,
+      "log_step": null,
+      "text_zh": "观察：视频 8.92–9.00 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 9.12 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第12次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 9.12,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第12次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 8.92–9.00 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第12次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.72,
+      "log_step": null,
+      "text_zh": "观察：视频 9.52–9.60 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 9.72 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第13次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 9.72,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第13次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 9.52–9.60 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第13次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.36,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.36,
+      "log_step": null,
+      "text_zh": "观察：视频 10.12–10.20 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 10.36 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第14次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 10.36,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第14次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 10.12–10.20 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第14次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.92,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.92,
+      "log_step": null,
+      "text_zh": "观察：视频 10.72–10.80 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 10.92 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第15次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 10.92,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第15次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 10.72–10.80 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第15次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [
+     [
+      11.56,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 11.56,
+      "log_step": null,
+      "text_zh": "观察：视频 11.32–11.44 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 11.56 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第16次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 11.56,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第16次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 11.32–11.44 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第16次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [
+     [
+      12.16,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 12.16,
+      "log_step": null,
+      "text_zh": "观察：视频 11.96–12.04 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 12.16 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第17次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 12.16,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第17次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 11.96–12.04 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第17次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [
+     [
+      12.76,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 12.76,
+      "log_step": null,
+      "text_zh": "观察：视频 12.56–12.64 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 12.76 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第18次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 12.76,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第18次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 12.56–12.64 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第18次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [
+     [
+      13.36,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 13.36,
+      "log_step": null,
+      "text_zh": "观察：视频 13.16–13.24 s，橙色小球在机器人右手所持红色拍面处由下落转为上升；视频 13.36 s 附近，球回到开局释放杯中球心的高度附近，随后继续升高。机器人保持右手持拍、双脚着地支撑；这是顺序检查录像得到的第19次反弹，时间约有一帧误差。"
+     },
+     {
+      "t": 13.36,
+      "log_step": 26,
+      "text_zh": "推断：上述反弹属于第19次有效颠球；整段录像可顺序对应20次触拍反弹，log step 26 中交付文件的独立重放裁判列出 bounce 1 至 bounce 20、bounces=20、success=true，且与本次评分器 reward.metric_bounces=20、stages_reached=20、deterministic=1 一致。有效速率比及毫秒接触门槛由裁判验收确认，未从像素估算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 13.16–13.24 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第19次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [
+     [
+      13.95,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 13.92,
+      "log_step": null,
+      "text_zh": "观察：视频 13.76–13.84 s，小球在机器人右手所持红色拍面处完成第20次由下落到上升的反转；视频 13.88–13.92 s 小球继续上升，最后可见帧仍略低于开局释放杯中球心高度。机器人仍右手持拍、双脚支撑，没有落球或跌倒的画面。"
+     },
+     {
+      "t": 13.95,
+      "log_step": 26,
+      "text_zh": "观察：log step 26 对交付文件进行独立重放，裁判输出 executed_rows=1395、steps=1395、time_s=13.95、success=true、failed=null、bounces=20，阶段包括 bounce 20；log step 27 再次重放成功且终态哈希相同，本次评分器亦记录 replay_success=1、deterministic=1、metric_bounces=20。"
+     },
+     {
+      "t": 13.95,
+      "log_step": 26,
+      "text_zh": "推断：第20次有效颠球在视频结束前达标。采用交付重放的控制步映射：log step 14 控制代码定义每步0.01秒，log step 25 将该控制器保存轨迹，log step 26 交付文件第1395控制步对应视频 1395×0.01=13.95 s，与视频总长13.96秒一致；视频 2.00 s 的放球及视频 13.76–13.84 s 的末次触拍已核对。此完成锚点比最后可见帧视频 13.92 s 晚0.03秒，并非声称看到了视频 13.95 s 的独立帧。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 13.76–13.84 s，机器人右手持红色球拍，拍面使橙色小球由下降转为上升。候选方式“击打”，接触词表没有对应专项，故用其他；握拍柄属于同一工具链，不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第20次颠球期间，机器人双脚保持接触地面，以腿和躯干维持站姿，未见迈步或跌倒。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": 20.0
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": 1.202
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.004
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人在视频约 2.28 s 首次触拍，随后保持双脚站立、右手持拍完成连续颠球，20个事件节点均达到，交付重放与本次评分器均确认20次有效计数及成功。视频逐帧可对应每次触拍和上升，第20次在视频 13.80 s 附近触拍后上升，完成锚点由交付重放第1395控制步定位至视频 13.95 s。没有状态日志，25 fps录像不能独立核实毫秒接触或0.98速率门槛，且末帧视频 13.92 s 比最终达标早约0.03秒；这些验收细节依据裁判结果确认。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "成功运行，评分器 stages_reached=20/20、metric_bounces=20，20个节点均达到且 final=true；抽查视频帧见持拍站立颠球，接触标注（手持工具 其他/击打、腿/躯干 支撑）与DAG和实际一致，无失败层，未发现需改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:55:11Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--xemhvcr/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--xemhvcr/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "fcab1116a76e86e86a476a65606fd43af53276b7e28d9df689599a714555bf94",
+   "video_sha256": "65c0d1f540557233bd79b706d3d70fed38d1c2adf76b1a5c45d70ab3b2d5c36c",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--xv8vjvd.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--xv8vjvd.json
new file mode 100644
index 0000000000000000000000000000000000000000..e001c6603d7088c7d854633071c91b566fb0e52f
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-privileged--xv8vjvd.json
@@ -0,0 +1,489 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-privileged--xv8vjvd",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.1,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 395,
+   "sim_steps": 395,
+   "horizon": 2000,
+   "episode_over": 1,
+   "progress": 0.1,
+   "stages_reached": 2,
+   "stages_total": 20,
+   "fell": 0,
+   "replay_s": 6.18,
+   "bad_trajectory": 0,
+   "metric_ball_height_m": 0.005,
+   "metric_bounces": 2.0,
+   "metric_grip_break_max_s": 0.014,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 83,
+   "agent_resets": 918,
+   "agent_replays": 3,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 2.026,
+   "peak_processes": 19,
+   "peak_disk_gb": 0.241,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--xv8vjvd/replay.mp4",
+  "duration_s": 3.96,
+  "fps": 25.0,
+  "frames": 99,
+  "sha256": "82f562fb2badd1e3f9928ec17961b525be73e382d39b14cac829e9279f27df6f"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 2,
+  "final": 2,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.44,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.44,
+      "log_step": null,
+      "text_zh": "观察：视频 2.16–2.28 s 橙色球落至机器人右手持有的红色拍面；视频 2.32–2.60 s 小球反向上升，视频约 2.44 s 回到开局释放杯中球心的高度，随后继续升高。"
+     },
+     {
+      "t": 2.44,
+      "log_step": 94,
+      "text_zh": "观察：log step 94 对最终保存的395行交付轨迹进行独立重放，裁判锁存 bounce 1、bounce 2；评分器 metric_bounces=2。推断：结合视频中首次触拍后的上升，将第1次有效颠球到达时刻定为视频约 2.44 s；速率及毫秒接触条件依据裁判计数确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：机器人右手持红色球拍，拍面击打橙色小球；候选方式为“击打”，接触词表没有对应项，依S16用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：机器人双脚在地面支撑身体，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.04,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": null,
+      "text_zh": "观察：视频 2.68–2.88 s 橙色球回落到红色拍面，视频 2.92–3.20 s 再次上升；视频约 3.04 s 已升至开局释放球高度以上。"
+     },
+     {
+      "t": 3.04,
+      "log_step": 98,
+      "text_zh": "观察：评分器 stages_reached=2、metric_bounces=2；log step 98 再次重放最终交付文件，得到 bounce 1、bounce 2，且终态哈希与log step 94一致。推断：这对应视频中前两次击球后的高弹起，第2次到达时刻约为视频 3.04 s。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：机器人右手持红色球拍，拍面击打橙色小球；候选方式为“击打”，接触词表没有对应项，依S16用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：机器人双脚在地面支撑身体，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.56,
+      "log_step": 94,
+      "text_zh": "观察：视频 3.32–3.60 s 橙色球回落经过机器人持拍手附近，但拍面没有将它再次向上击起；视频 3.64–3.92 s 球向画面左下方落到近地处，机器人仍站立并持拍。log step 94 对最终交付轨迹的重放报告“the ball touched the floor”，仅完成 bounce 1、bounce 2；没有第3次有效颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 3.32–3.60 s 机器人继续右手持红色球拍等待并尝试击打回落球，球从拍面旁落下，未形成有效回弹；不能从25 fps画面确认是否短暂擦到拍缘。按实际击打尝试标注，候选方式“击打”，词表无此项依S16用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：机器人双脚在地面支撑身体，同时右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第4次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第5次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第6次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第7次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第8次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第9次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第10次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第11次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第12次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第13次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第14次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第15次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第16次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第17次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第18次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第19次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.92,
+      "log_step": 94,
+      "text_zh": "观察：本次视频 0.00–3.92 s 只有两次击球后的高弹起，第三次回落后小球从拍面旁落下，未出现第20次有效颠球；log step 94 对最终交付文件的重放仅列出 bounce 1、bounce 2，评分器 metric_bounces=2、stages_reached=2。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": 2.0
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": 0.005
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.014
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_3",
+   "t": 3.56,
+   "log_step": 94,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 3.32–3.60 s 机器人正在用右手所持红色球拍接续颠球，橙色球回落时从拍面旁通过，没有第三次向上回弹；视频 3.64–3.92 s 小球继续向地面落下。log step 94、98 对最终395行交付轨迹的重放均由裁判报告球触地，并只锁存前两次有效颠球；第3次颠球从未完成。"
+  },
+  "unmet": {
+   "bounce_1": true,
+   "bounce_2": true,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 3.56,
+    "log_step": null,
+    "evidence_zh": "观察：视频 3.32–3.60 s 橙色球从高处回落，机器人持拍手与拍面未能跟上球的回落位置，球由拍面旁向画面左下方落下；视频 3.64–3.92 s 没有再次向上的回弹。推断：这是对运动球的击打未能衔接，属动态目标的跟随失败。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "source_zh": "视频 / 日志",
+    "evidence_zh": "推断：视频 2.28–3.20 s 证明机器人已能持拍完成前两次击球及达标上升，但视频 3.32–3.60 s 球与拍面的相对位置偏离，持拍动作未能修正到下一次击球位置，故归为连续颠球的控制失败。log step 88、92、94 显示最终交付沿用ctrl10的持拍、拍面调平及击球控制，独立重放仍只有2次；log step 101 中agent将其解释为拍面残余倾斜及握柄微滑导致横向漂移，其中微滑未获独立状态证据确认，不据此细分为接触设计或参数问题。"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志。25 fps场地录像不能独立测定0.02秒触球间隔、0.98速度比或握柄微滑；前两次有效性以最终交付重放的裁判阶段和评分器计数确认，到达时间取视频可见上升的近似时刻。第三次球经过持拍手附近时，无法可靠区分完全漏接与短暂擦过拍缘，但未形成有效上升及随后触地有视频和重放结果支持；控制根因保留大类。"
+ },
+ "summary_zh": "机器人右手持拍并保持站立，在视频约 2.44 s、3.04 s 完成前两次有效颠球，最终进度为2/20。视频约 3.56 s 第三次回落未获有效击打，球从拍面旁落下，最终交付的两次一致重放均报告球触地；断点为第3次颠球，现象为动态时机，根因归于控制。缺少逐帧状态日志，不能确认短暂擦边或agent所述握柄微滑，精确接触和速率阈值依裁判计数确认。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "逐帧核对：球 2.0 s 释放，约 2.2–2.3 s 触拍后上升，约 2.5 s 超过释放高度；约 2.85 s 回落再击起，约 3.04 s 已高于释放高度；约 3.44 s 第三次回落到手附近后消失、无再次上升。bounce_1/2 为真、bounce_3 起未达成，断点 bounce_3、操作中，与评分器 stages_reached=2 一致。现象与根因保守可接受。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:55:32Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--xv8vjvd/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-privileged--xv8vjvd/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "09141d07727a8fcd59210094c5768f0816593d2f820d29fa5d1745fb13f76210",
+   "video_sha256": "82f562fb2badd1e3f9928ec17961b525be73e382d39b14cac829e9279f27df6f",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--9ywbdae.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--9ywbdae.json
new file mode 100644
index 0000000000000000000000000000000000000000..3d737f3519f433617c6525b8f6629553ccb2d957
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--9ywbdae.json
@@ -0,0 +1,830 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard--9ywbdae",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "live_success": 1,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 1252,
+   "sim_steps": 1723,
+   "calls": 1232,
+   "rows": 1723,
+   "resets": 2,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 684.7,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.198,
+   "peak_processes": 12,
+   "peak_disk_gb": 0.346,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard--9ywbdae/replay.mp4",
+  "duration_s": 12.56,
+  "fps": 25.0,
+  "frames": 314,
+  "sha256": "e8b97bf891356191a5c14d9ffab9fc4786ed1a3a99bad934871b5530301c5f24"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 20,
+  "final": 20,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.36,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.36,
+      "log_step": null,
+      "text_zh": "观察：视频约 2.20 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 2.36 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第1个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 2.36,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第1次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 2.20 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第1次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.88,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.88,
+      "log_step": null,
+      "text_zh": "观察：视频约 2.72 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 2.88 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第2个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 2.88,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第2次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 2.72 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第2次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.44,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.44,
+      "log_step": null,
+      "text_zh": "观察：视频约 3.28 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 3.44 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第3个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 3.44,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第3次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 3.28 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第3次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.96,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.96,
+      "log_step": null,
+      "text_zh": "观察：视频约 3.80 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 3.96 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第4个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 3.96,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第4次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 3.80 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第4次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.48,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.48,
+      "log_step": null,
+      "text_zh": "观察：视频约 4.32 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 4.48 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第5个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 4.48,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第5次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 4.32 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第5次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.04,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.04,
+      "log_step": null,
+      "text_zh": "观察：视频约 4.88 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 5.04 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第6个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 5.04,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第6次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 4.88 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第6次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.56,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.56,
+      "log_step": null,
+      "text_zh": "观察：视频约 5.40 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 5.56 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第7个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 5.56,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第7次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 5.40 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第7次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.12,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.12,
+      "log_step": null,
+      "text_zh": "观察：视频约 5.96 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 6.12 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第8个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 6.12,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第8次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 5.96 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第8次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.64,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.64,
+      "log_step": null,
+      "text_zh": "观察：视频约 6.48 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 6.64 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第9个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 6.64,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第9次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 6.48 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第9次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.2,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.2,
+      "log_step": null,
+      "text_zh": "观察：视频约 7.04 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 7.20 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第10个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 7.2,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第10次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 7.04 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第10次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.72,
+      "log_step": null,
+      "text_zh": "观察：视频约 7.56 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 7.72 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第11个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 7.72,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第11次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 7.56 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第11次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.24,
+      "log_step": null,
+      "text_zh": "观察：视频约 8.08 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 8.24 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第12个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 8.24,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第12次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 8.08 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第12次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.8,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.8,
+      "log_step": null,
+      "text_zh": "观察：视频约 8.64 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 8.80 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第13个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 8.8,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第13次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 8.64 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第13次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.36,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.36,
+      "log_step": null,
+      "text_zh": "观察：视频约 9.20 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 9.36 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第14个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 9.36,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第14次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 9.20 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第14次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.88,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.88,
+      "log_step": null,
+      "text_zh": "观察：视频约 9.72 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 9.88 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第15个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 9.88,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第15次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 9.72 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第15次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.44,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.44,
+      "log_step": null,
+      "text_zh": "观察：视频约 10.28 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 10.44 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第16个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 10.44,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第16次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 10.28 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第16次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.96,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.96,
+      "log_step": null,
+      "text_zh": "观察：视频约 10.80 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 10.96 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第17个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 10.96,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第17次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 10.80 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第17次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [
+     [
+      11.52,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 11.52,
+      "log_step": null,
+      "text_zh": "观察：视频约 11.36 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 11.52 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第18个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 11.52,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第18次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 11.36 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第18次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [
+     [
+      12.04,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 12.04,
+      "log_step": null,
+      "text_zh": "观察：视频约 11.88 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 12.04 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第19个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 12.04,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第19次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 11.88 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第19次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [
+     [
+      12.52,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 12.52,
+      "log_step": null,
+      "text_zh": "观察：视频约 12.40 s 橙色球下降至机器人右手所持红色拍面附近，随后向上反弹；视频 12.52 s 球已明显离拍并上升至开局释放高度附近或更高。前后连续帧显示这是本局第20个完整触拍后上升周期；机器人双脚着地，右手继续持拍。"
+     },
+     {
+      "t": 12.52,
+      "log_step": 38,
+      "text_zh": "推断：视频逐帧共见20个触拍后上升周期，最后一局服务裁判在 log step 38 返回 DONE success、stages_reached=20、stages_total=20，log step 39 确认 episode=3 已结束且 reason=success，因此将该周期对应为第20次有效颠球并锁存。速度比与毫秒接触阈值由裁判成功结果支持，视频达成时刻约有0.1至0.2秒估计误差。"
+     },
+     {
+      "t": 12.52,
+      "log_step": 39,
+      "text_zh": "推断：log step 3 的服务规格给出 control_dt=0.01 s；log step 39 确认最后受评局 episode=3 的 control_steps=1252，即仿真12.52秒，核对视频12.40至12.52 s 最后一次离拍上升及终止帧一致。这里使用受评局控制步数核对视频末端，未将agent工作时间转换为视频时间。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 12.40 s 机器人右手持红色球拍，以拍面将下落橙色乒乓球击起。候选方式为“击打”；现有接触方式无击打项，故填其他；同一工具链不另重复记录手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：第20次颠球期间机器人保持站立，双脚接触地面支撑本体，右臂持拍完成击球。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": 20
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": null,
+ "summary_zh": "受评录像对应两次重置后的第3局：机器人双脚站立、右手持拍，在视频约2.20至12.52 s完成20个触拍后上升周期，20个事件节点最终全部达成。log step 38 的服务裁判返回success及20/20阶段，log step 39确认第3局在1252个控制步后成功结束，原生bounces=20据该裁判阶段数填写；会话中前两局失败不属于本次受评轨迹。未提供状态日志，逐次时刻由25 fps连续帧估计，0.98速度比及毫秒级接触规则依成功裁判验收确认，ball_height_m和grip_break_max_s未在评分结果中报告故留null；参考成功录像仅用于比较可见形态。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "成功局（裁判 20/20），20 个事件节点均达成，时间与 25 fps 视频一致，接触（手持工具·其他/击打 + 腿/躯干·支撑）符合实际，无失败层，未发现需修改之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:56:55Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--9ywbdae/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--9ywbdae/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "ad0d4541f49aff5af22718ec767c491ce608e2d60166da06e4103a0a82f46765",
+   "video_sha256": "e8b97bf891356191a5c14d9ffab9fc4786ed1a3a99bad934871b5530301c5f24",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--as9pd4b.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--as9pd4b.json
new file mode 100644
index 0000000000000000000000000000000000000000..bae2e577bf0e822d9f69b7ee5e6f235af4c9ebfe
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--as9pd4b.json
@@ -0,0 +1,462 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard--as9pd4b",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 301,
+   "sim_steps": 1480,
+   "calls": 64,
+   "rows": 1480,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2068.3,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.451,
+   "peak_processes": 10,
+   "peak_disk_gb": 1.13,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard--as9pd4b/replay.mp4",
+  "duration_s": 3.04,
+  "fps": 25.0,
+  "frames": 76,
+  "sha256": "8572a99e1e1f433301efb39c0aaffedcdefe75c10aa661c68ab14816a3cc4731"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.36,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.36,
+      "log_step": 174,
+      "text_zh": "观察：视频2.20–2.28 s中，机器人右手所持红色拍面接住下落橙球并将球击起；视频2.36–2.52 s小球升回并超过开局释放杯中的球心高度。log step 174中最后一局服务返回stages_reached=1、stages_total=20；该字段是仿真服务裁判的累计有效颠球阶段，补充确认画面不能测出的速率与接触时长条件。首次达标锚点取视频约2.36 s，误差约0.04 s。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约2.24 s，机器人右手持红色球拍，用拍面击起橙色小球；候选方式“击打”，现有接触方式词表无击打项，故用其他。握拍柄不重复计接触链。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.00–3.00 s中，机器人双脚接触地面支撑身体，右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.76,
+      "log_step": 174,
+      "text_zh": "观察：视频2.56–2.72 s小球从首次反弹高点下降并向机器人侧偏移；视频2.76–2.80 s小球从红色拍面的靠身体一侧漏下，没有再次向上反弹，视频2.84–3.00 s继续降至地面。log step 174最后一局服务返回球触地、阶段1/20，故第2次有效颠球未达成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频2.56–2.80 s，机器人继续用右手持红色球拍上下运动，尝试第二次击打，但拍面未跟上偏移的小球，小球从拍旁漏下；此处记录实际持拍击打尝试，不表示拍面实际再次触球。候选方式“击打”，词表无对应项，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频2.56–3.00 s，机器人双脚持续在地面支撑持拍动作，未见跌倒。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第3次未达到，其前置第2次亦未完成。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第4次未达到，其前置第3次亦未完成。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第5次未达到，其前置第4次亦未完成。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第6次未达到，其前置第5次亦未完成。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第7次未达到，其前置第6次亦未完成。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第8次未达到，其前置第7次亦未完成。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第9次未达到，其前置第8次亦未完成。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第10次未达到，其前置第9次亦未完成。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第11次未达到，其前置第10次亦未完成。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第12次未达到，其前置第11次亦未完成。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第13次未达到，其前置第12次亦未完成。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第14次未达到，其前置第13次亦未完成。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第15次未达到，其前置第14次亦未完成。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第16次未达到，其前置第15次亦未完成。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第17次未达到，其前置第16次亦未完成。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第18次未达到，其前置第17次亦未完成。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第19次未达到，其前置第18次亦未完成。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.0,
+      "log_step": 174,
+      "text_zh": "观察：视频2.76–3.00 s，小球漏过拍面后持续下落至地面，未发生后续颠球；log step 174最后一局服务裁判返回球触地、阶段1/20。推断：累计有效颠球仅1次，故第20次未达到，其前置第19次亦未完成。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "value": 1,
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better"
+  },
+  {
+   "name": "ball_height_m",
+   "value": null,
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better"
+  },
+  {
+   "name": "grip_break_max_s",
+   "value": null,
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_2",
+   "t": 2.76,
+   "log_step": 174,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约2.36 s首次反弹升回释放高度后，机器人继续持拍上下运动；视频2.76–2.80 s小球在第二次下降中漏过拍面，随后落地，未再恢复颠球。log step 174服务报告最后一局球触地、阶段1/20；log step 180确认episode=6、control_steps=301且done=true，之后没有新受评回合。"
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 2.76,
+    "log_step": 174,
+    "evidence_zh": "观察：视频2.56–2.80 s中，运动中的小球下降并横向偏移，机器人右手球拍仍在原位置附近上下运动，未跟上回落小球，球从拍面靠身体一侧漏下；视频2.84–3.00 s中小球继续降至地面。此为第二次击球对运动目标的跟随失败。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "反馈控制",
+    "evidence_zh": "观察：log step 174是最后一局episode 6的执行代码及输出，脚本只有在仿真时刻大于2.35 s且检测球高于1.06米时才更新横向off；输出在仿真2.31 s尚未到时间门槛，仿真2.52 s未检测到球，仿真2.73 s检测球高度为1.007米，off始终为(0,0)。视频2.56–2.80 s同时显示小球偏离而拍面未横向追随。推断：该反馈更新条件使第二次触球前的横向纠偏未执行，机器人沿原位置上下挥拍，导致漏接。",
+    "source_zh": "日志"
+   }
+  ],
+  "gap_zh": "没有逐帧状态或毫秒级接触记录，首次有效计数由视频反弹与最后一局服务裁判阶段1/20共同确认，无法从25 fps画面独立测量0.98速度比及0.02秒触球时长。评分器和服务输出未报告原生ball_height_m、grip_break_max_s数值，保留null；不把agent自行估计的位置或末尾自述的精确漂移量当作状态真值。",
+  "unmet": {
+   "bounce_1": true,
+   "bounce_2": false,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  }
+ },
+ "summary_zh": "受评轨迹为最后一局episode 6：机器人右手持拍，在视频约2.36 s完成第1次有效颠球，最终达成1/20个节点。第二次回落时拍面未跟上小球横向偏移，视频约2.76 s漏接，随后球落地；最后一局控制脚本的横向反馈条件未触发，导致纠偏未执行。没有状态日志，毫秒接触和速率门槛依靠该局服务裁判阶段确认，另两项原生数值未报告。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标签与证据一致：视频约2.36 s首次有效颠球，随后球在约2.76 s漏过拍面落地，断点 bounce_2、动态时机现象均有log step 174（仿真2.73 s球x=0.283而拍面x=0.361，终局球触地、阶段1/20）支撑；根因反馈控制有代码和输出支撑，可接受。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:58:57Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--as9pd4b/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--as9pd4b/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "4a2dfa4599c4a9bb2acc8d03a5cffc71487130f6d810efe43021c6739506207f",
+   "video_sha256": "8572a99e1e1f433301efb39c0aaffedcdefe75c10aa661c68ab14816a3cc4731",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--cm5vyfg.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--cm5vyfg.json
new file mode 100644
index 0000000000000000000000000000000000000000..17a6dba35341306ce32f99f87c39998c757124cc
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--cm5vyfg.json
@@ -0,0 +1,455 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard--cm5vyfg",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 252,
+   "sim_steps": 1196,
+   "calls": 31,
+   "rows": 1196,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2812.9,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 4.996,
+   "peak_processes": 12,
+   "peak_disk_gb": 2.53,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard--cm5vyfg/replay.mp4",
+  "duration_s": 2.56,
+  "fps": 25.0,
+  "frames": 64,
+  "sha256": "4d1b55747fe7960e15bb95627fc679a6ab2c3f616175121c562d542e90b7df90"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.12,
+      "log_step": 172,
+      "text_zh": "观察：视频0.64–1.12 s机器人把右手所持红色球拍从身侧抬到身前；视频1.52–1.92 s继续转平拍面。log step 172是最后一次重置后的抬拍动作，随后log step 175–177继续同一局，未再重置。"
+     },
+     {
+      "t": 2.24,
+      "log_step": 177,
+      "text_zh": "观察：视频2.12–2.36 s橙球由释放杯落下，从红色拍面旁持续下落，未被拍面击起；视频2.48–2.52 s球已到近地处，机器人正向前倾倒。log step 177中机器人服务返回“the ball touched the floor”，与本次录像一致；未见任何一次达标上升。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.64–2.28 s机器人双脚接地支撑身体，同时右手持拍调整位置；视频2.32–2.52 s支撑姿态失稳并向前倾倒。球从拍面旁落过，没有观察到球拍与球接触，故不把计划中的拍面击打填成实际接触；持拍不重复计入工具作用链。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第2次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第3次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第4次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第5次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第6次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第7次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第8次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第9次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第10次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第11次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第12次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第13次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第14次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第15次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第16次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第17次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第18次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第19次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 177,
+      "text_zh": "推断：视频2.12–2.52 s小球连续下落至近地处，无首次拍球反弹；log step 177中机器人服务终止该局并返回“the ball touched the floor”。第1次有效颠球尚未完成，故按顺序累计的第20次有效颠球也未到达，后续步骤未开始。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_1",
+   "t": 2.24,
+   "log_step": 177,
+   "where": "操作中",
+   "evidence_zh": "观察：视频0.64–1.92 s机器人已抬拍并转动拍面，开始为首球作准备；视频2.12–2.36 s小球从拍面旁漏下，机器人没有将球击回释放高度，此后直到视频2.52 s结束也没有修复。log step 177记录的是继续最后一局的控制调用，机器人服务返回“the ball touched the floor”；log step 180确认episode=6、done=True、resets_left=0。失败所在步骤为第1次颠球。"
+  },
+  "unmet": {
+   "bounce_1": false,
+   "bounce_2": false,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 2.24,
+    "log_step": 177,
+    "evidence_zh": "观察：视频2.12–2.36 s运动中的橙球从杯下落，经机器人身侧落到拍面以下，右手拍面未跟到落球轨迹下方，也没有将球反弹起来；log step 177的服务结果为球触地。推断：首球下落期间的持拍跟随未与球的运动配合，属于动态目标漏击。"
+   },
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 2.4,
+    "log_step": null,
+    "evidence_zh": "观察：视频1.76–2.28 s机器人整体逐渐向前倾斜，视频2.32–2.52 s倾斜继续加剧，脚下未出现有效恢复步，末帧躯干已接近水平，仍处于向前倾倒过程。画面结束前未清楚显示躯干触地。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 2.24,
+    "log_step": 177,
+    "source_zh": "日志",
+    "note_zh": "控制脚本时钟索引错误，击球控制相位未与实际回合进度同步。",
+    "evidence_zh": "观察：log step 9打印的服务proprio_layout明确列出140=cmd_vy、142=time_s；log step 172的脚本修改却将击球循环时间设为float(pr3[140])，并从pr4[140]更新。log step 175继续该脚本、log step 177修复q1后执行，同一次运行先打印预计首次触拍时刻2.188，随后连续两次打印“chunk 0.00-0.42”，第二块收到机器人服务的球触地结果。视频2.12–2.36 s也显示漏球。推断：脚本把侧向速度指令读成回合时间，使击球相位与真实下落过程错开；证据支持控制实现错误，未证明具体的动力学或逆解子类。"
+   }
+  ],
+  "gap_zh": "没有状态日志，也没有本局原生颠球数、球高及持拍中断时长的评分器指标，无法复核毫秒级触球或0.98速率门槛；三个native_metrics按未报告填null。视频足以判定未出现首次反弹。前倾失稳的独立动力学原因无法由现有记录细分，不将agent关于重置预算的自述当作根因。"
+ },
+ "summary_zh": "受评的是最后一局：机器人抬起并转平右手球拍，但视频2.12–2.36 s小球从拍面旁漏下，无一次有效颠球，最终完成0/20个节点。第1次颠球在操作中失败，随后机器人明显向前倾倒；日志证实脚本把cmd_vy索引140当作时间，而实际时间索引为142，导致击球相位失配。缺少状态日志及原生接触、速率指标，不能细分失稳的动力学原因。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频仅2.56 s，最后一局球从拍面旁落下、无有效颠球，机器人向前倾倒；全部节点未达成、断点bounce_1、动态时机+平衡及控制类根因（脚本把cmd_vy索引140当作time_s，日志layout中time_s=142）均有帧和日志支持，未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:52:24Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--cm5vyfg/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--cm5vyfg/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "926f93c6686937578c02796cafa715e961f71ddd8b1aab7e8cc848b6c72adfae",
+   "video_sha256": "4d1b55747fe7960e15bb95627fc679a6ab2c3f616175121c562d542e90b7df90",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--ingeat5.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--ingeat5.json
new file mode 100644
index 0000000000000000000000000000000000000000..415eef55563939b4a661dadc8f2f6538cec7719c
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--ingeat5.json
@@ -0,0 +1,448 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard--ingeat5",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 226,
+   "sim_steps": 1074,
+   "calls": 27,
+   "rows": 639,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 838.4,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.305,
+   "peak_processes": 11,
+   "peak_disk_gb": 3.093,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard--ingeat5/replay.mp4",
+  "duration_s": 2.28,
+  "fps": 25.0,
+  "frames": 57,
+  "sha256": "5089681a9ee1fc75b4a0c4c37839f71dfa132f6f8fd79364040e06062277e372"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 中机器人双脚支撑，右手将红色球拍由身侧移至身前；视频 2.08–2.24 s 橙球自释放器下落至右手/拍柄邻近区域，未见被拍面击回释放高度。log step 83 的机器人实时服务返回 episode=6、done=True、stages_reached=0、stages_total=20，并报告球触及拍面以外的物体；该末局与视频动作及终止位置一致，第1次有效颠球未完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 0.00–2.24 s 机器人右手持续持球拍并摆拍迎接橙球；候选方式为“击打”（尝试），词表无对应方式。未观察到拍面成功击球，视频末段球进入右手/拍柄邻近区域；不把持拍本身重复记为工具对球的抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–2.24 s 机器人双脚接触地面，支撑站立并配合右臂持拍迎球。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第2次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第3次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第4次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第5次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第6次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第7次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第8次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第9次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第10次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第11次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第12次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第13次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第14次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第15次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第16次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第17次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第18次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第19次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": 83,
+      "text_zh": "观察：视频 0.00–2.24 s 仅出现放球前摆拍和首次下落迎球，没有后续反复颠球；log step 83 的末局实时裁判记录为 stages_reached=0/20。推断：按有效颠球的累计顺序，第20次颠球未达到，且未开始独立尝试。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_1",
+   "t": 2.24,
+   "log_step": 83,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 1.52–2.24 s 机器人把右手球拍移到身前并做短促迎球动作；视频 2.08–2.24 s 橙球落到右手/拍柄邻近区域，尚未出现向上回弹，录像即结束。log step 83 的末局机器人服务记录在 control_steps=226、仿真 2.26 s 以“the ball touched something other than the paddle's blade”终止，stages_reached=0/20；正在进行的第1次颠球未修复。"
+  },
+  "unmet": {
+   "bounce_1": false,
+   "bounce_2": false,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 2.24,
+    "log_step": 83,
+    "evidence_zh": "观察：视频 2.08–2.24 s 球持续下落，机器人右手持拍做小幅下压后回摆，未能让拍面及时截住下落球并击起；球进入手/柄邻近区域。log step 83 的末局实时裁判确认球接触拍面之外物体并终止，计数仍为0/20。推断：这是迎接运动球的失配，标动态时机。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "抓取和接触设计",
+    "alt_feasible": "已验证有",
+    "t": 2.24,
+    "log_step": 83,
+    "source_zh": "视频 / 日志",
+    "evidence_zh": "推断：所选持拍迎球姿态未把合法拍面接触区域可靠置于落球路径，球先到达右手/拍柄邻近区域。依据是视频 1.76–2.24 s 的拍、手、球相对位置，以及 log step 78 固定右臂目标后，log step 83 仅用肩、肘各0.08弧度变化执行短行程下压回摆，随后服务报告非拍面触球；不能据此细分为跟踪或估计误差。成功参考 inputs/oracle/replay.mp4 的参考视频 2.00–3.00 s 可见持拍拍面将球反复击起，证明存在合法替代接触布局，仅作可行性参考，不作为本局进度证据。"
+   }
+  ],
+  "gap_zh": "观察：inputs/ 内没有状态日志；视频最后可见帧为视频 2.24 s，log step 83 报告终止于仿真 2.26 s。末段单视角不能确定非法接触具体发生在右手还是拍柄，也不能测定毫秒级接触和速率比；根因中的接触布局为结合画面与动作日志的推断。run.json 的 reward 未报告 bounces、ball_height_m、grip_break_max_s 原生指标值，故 native_metrics 对应值留 null；有效计数0/20来自末局服务的阶段结果。"
+ },
+ "summary_zh": "观察：评分视频中的最后一局只完成持拍摆位和首次落球迎击，log step 83 的实时裁判记录有效颠球0/20，20个节点均未达到。推断：第1次颠球在视频约 2.24 s 失配，持拍布局使下落球进入右手/拍柄邻近区域，服务以非拍面触球终止；缺少接触状态日志，无法确定碰撞的具体部件。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "20个节点均未达到、断点 bounce_1、时间锚点与 log step 83 的服务读数（stages_reached=0/20，球触及拍面以外物体，仿真2.26 s）及视频末帧一致；现象“动态时机”与根因推断标注合理，未发现需要改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:01:43Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--ingeat5/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--ingeat5/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "ec78ea3994074f872ba52e08eb3f961cfb2d6c7fd2574dcfc503e207dbb34002",
+   "video_sha256": "5089681a9ee1fc75b4a0c4c37839f71dfa132f6f8fd79364040e06062277e372",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--jzyrz8f.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--jzyrz8f.json
new file mode 100644
index 0000000000000000000000000000000000000000..b88554329a6f09df7c1c2bb174eb752fa7ea642f
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--jzyrz8f.json
@@ -0,0 +1,474 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard--jzyrz8f",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 315,
+   "sim_steps": 1721,
+   "calls": 81,
+   "rows": 1721,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 1127.8,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.566,
+   "peak_processes": 6,
+   "peak_disk_gb": 0.376,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard--jzyrz8f/replay.mp4",
+  "duration_s": 3.16,
+  "fps": 25.0,
+  "frames": 79,
+  "sha256": "395a6eacd973285185bf9beda55bad3548e23ad75d9dfedf993695b1ebebf3ae"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.48,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.32,
+      "log_step": 138,
+      "text_zh": "观察：视频 2.28–2.36 s，机器人右手持红色球拍接触下落的橙色小球，小球随即转为上升；视频 2.48–2.64 s，小球升至开局释放杯中球心高度以上。机器人双脚始终支撑地面。"
+     },
+     {
+      "t": 2.48,
+      "log_step": 141,
+      "text_zh": "观察：log step 138 记录最后一次重置后执行的第6回合；log step 141 的服务裁判元数据为 episode=6、stages_reached=1、stages_total=20、outcome=the ball touched the floor。推断：该回合与视频的单次有效反弹及随后落地一致，第1次有效颠球约在视频 2.48 s 达标并锁存；精确速度比和毫秒接触门槛由裁判计数确认，时间按画面估计。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 2.32 s，机器人右手所持球拍的红色拍面将橙色小球击起；候选方式“击打”，现有接触方式无对应项，故用其他；不重复记录握拍柄接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–3.12 s，机器人双脚接触地面并支撑身体完成持拍击球，未见跌倒。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.96,
+      "log_step": 138,
+      "text_zh": "观察：视频 2.64–2.96 s，橙色小球持续向画面左侧偏移并下降，机器人球拍仍在原持拍区域附近；视频 2.92–3.00 s 有上抬挥拍动作，但小球从拍旁下落，没有第二次反弹。log step 138 的脚本打印触发 Stroke 2，而服务最终只报告 stages_reached=1/20。"
+     },
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 3.00–3.12 s，小球降至地面附近；log step 141 的最后回合服务结果为 the ball touched the floor、stages_reached=1。推断：机器人已尝试第二次颠球但未完成；脚本的 Stroke 2 是动作触发记录，不是裁判有效颠球计数。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：视频 2.64–3.12 s，机器人右手持续包握拍柄并挥动球拍尝试接续颠球；小球从拍旁下落，未见第二次球与拍面接触，因此此处只记录实际存在的右手与拍柄抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.64–3.12 s，机器人双脚留在地面，支撑持拍和挥拍动作。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第3次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第4次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第5次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第6次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第7次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第8次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第9次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第10次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第11次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第12次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第13次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第14次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第15次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第16次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第17次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第18次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第19次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.12,
+      "log_step": 141,
+      "text_zh": "观察：视频 0.00–3.12 s 仅见一次球被拍击起，随后球偏离球拍并下落；log step 141 的第6回合服务裁判结果为 stages_reached=1/20、the ball touched the floor。推断：累计第20次有效颠球未达成，且第2次漏接后回合结束，机器人未开始该次颠球。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "value": 1,
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better"
+  },
+  {
+   "name": "ball_height_m",
+   "value": null,
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better"
+  },
+  {
+   "name": "grip_break_max_s",
+   "value": null,
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_2",
+   "t": 2.96,
+   "log_step": 138,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 2.64–2.96 s，小球在首次反弹后向画面左侧偏飞并下降，机器人未将球拍移到球下方；视频 2.92–3.00 s 球拍上抬但挥空，随后小球继续降至地面附近。log step 138、141 的第6回合裁判结果为仅完成1/20次，因球触地失败。推断：首次未修复失败发生在接续第2次颠球的操作中。"
+  },
+  "unmet": {
+   "bounce_1": true,
+   "bounce_2": false,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 2.96,
+    "log_step": 138,
+    "evidence_zh": "观察：视频 2.72–2.96 s，下降中的橙色小球逐渐偏到球拍的画面左侧；视频 2.92–3.00 s，机器人在原持拍区域上抬球拍，小球从旁边落过且未反弹。推断：机器人没有跟上运动中的球，第二次挥拍漏接；不将这一漏接重复标为另一接触现象。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "反馈控制",
+    "log_step": 137,
+    "t": 2.96,
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "观察：log step 137 的最终控制脚本以球的横向像素偏差 du 计算限幅为±0.02 rad的 tilt_y，再将 dq_tilt_y*tilt_y 加到固定准备姿态或预先计算的挥拍轨迹；没有根据本次球的横向落点重新生成平移接球轨迹。log step 138 显示脚本已触发第二次挥拍；视频 2.64–3.00 s 显示小球持续偏离，球拍仍停留在原区域附近。推断：实际闭环修正未能让球拍跟随首次反弹后的球路，导致第二次挥拍漏接；不据此推定视觉识别错误或具体关节参数错误。"
+   }
+  ],
+  "gap_zh": "未提供逐帧状态日志；视频25 fps不能独立复核0.02 s接触时长及0.98速度比，第1次有效颠球由最终回合的服务裁判计数确认，达标时间约取视频 2.48 s。缺少末端姿态及球三维轨迹，无法进一步确认首次偏飞来自拍面倾角、手内转动还是其他接触细节；原生 ball_height_m 与 grip_break_max_s 未报告，留空。"
+ },
+ "summary_zh": "受评分的是 log step 138 最后重置后执行的第6回合；视频约 2.32 s 右手持拍击球，约视频 2.48 s 球升至释放高度，服务裁判确认完成第1次有效颠球，终态进度1/20。首次反弹后球向画面左侧偏飞，闭环持拍修正未跟上下降球路，在视频约 2.96 s 第二次挥拍漏接，随后裁判因球触地判负。无逐帧状态日志，首次达标时刻取画面近似值，精确触球和速度门槛依裁判计数确认；颠球数1由服务裁判 stages_reached 换读，其余原生指标未报告。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "节点、时间、断点与现象均与视频及服务裁判读数（stages_reached=1/20，球触地）一致：视频约2.3 s击球、2.5 s球升起，随后球向左偏离，球拍原地未跟上，第二次挥拍（log step 138，sim 2.9 s）漏接。动态时机（没跟上）与反馈控制可接受；未发现需改动的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:49:05Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--jzyrz8f/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--jzyrz8f/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "a7bef384aecb745a67dd3ba763b40a659b193b65a4a387d632efc9f16c5e89c6",
+   "video_sha256": "395a6eacd973285185bf9beda55bad3548e23ad75d9dfedf993695b1ebebf3ae",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--nx7ih2g.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--nx7ih2g.json
new file mode 100644
index 0000000000000000000000000000000000000000..8fcbd722459f71ed8daafb5fd53bc4c0d4e321c5
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--nx7ih2g.json
@@ -0,0 +1,511 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard--nx7ih2g",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 419,
+   "sim_steps": 3988,
+   "calls": 831,
+   "rows": 2608,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2030.9,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.445,
+   "peak_processes": 9,
+   "peak_disk_gb": 2.051,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard--nx7ih2g/replay.mp4",
+  "duration_s": 4.2,
+  "fps": 25.0,
+  "frames": 105,
+  "sha256": "11cd73506422e75131fd3f514768fd4d985d476b2038b3e94591871f65044ccd"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 3,
+  "final": 3,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.32,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.32,
+      "log_step": 93,
+      "text_zh": "观察：视频约 2.20 s，机器人右手所持红色拍面使橙球由下降转为上升；视频约 2.32–2.44 s，球升回开局释放杯高度附近或更高，为第1次可见反弹。log step 93 的机器人服务终局结果为 episode=6、stages_reached=3、stages_total=20，与视频中的三次反弹一致。"
+     },
+     {
+      "t": 2.32,
+      "log_step": 93,
+      "text_zh": "推断：结合三次可见反弹与本局裁判确认的3次有效计数，此次为有效颠球；以球上升达标的可见帧为到达时刻，时间约有0.1秒误差。0.98速度比与毫秒级触拍验收依裁判总数确认，未由录像测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 2.20 s，机器人右手持球拍，红色拍面击打橙色球；候选方式“击打”，现有接触方式词表无此项，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频约 2.20–2.44 s，机器人双脚接地支撑身体，右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.76,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.76,
+      "log_step": 93,
+      "text_zh": "观察：视频约 2.64 s，机器人右手所持红色拍面使橙球由下降转为上升；视频约 2.76–2.88 s，球升回开局释放杯高度附近或更高，为第2次可见反弹。log step 93 的机器人服务终局结果为 episode=6、stages_reached=3、stages_total=20，与视频中的三次反弹一致。"
+     },
+     {
+      "t": 2.76,
+      "log_step": 93,
+      "text_zh": "推断：结合三次可见反弹与本局裁判确认的3次有效计数，此次为有效颠球；以球上升达标的可见帧为到达时刻，时间约有0.1秒误差。0.98速度比与毫秒级触拍验收依裁判总数确认，未由录像测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 2.64 s，机器人右手持球拍，红色拍面击打橙色球；候选方式“击打”，现有接触方式词表无此项，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频约 2.64–2.88 s，机器人双脚接地支撑身体，右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.24,
+      "log_step": 93,
+      "text_zh": "观察：视频约 3.12 s，机器人右手所持红色拍面使橙球由下降转为上升；视频约 3.24–3.36 s，球升回开局释放杯高度附近或更高，为第3次可见反弹。log step 93 的机器人服务终局结果为 episode=6、stages_reached=3、stages_total=20，与视频中的三次反弹一致。"
+     },
+     {
+      "t": 3.24,
+      "log_step": 93,
+      "text_zh": "推断：结合三次可见反弹与本局裁判确认的3次有效计数，此次为有效颠球；以球上升达标的可见帧为到达时刻，时间约有0.1秒误差。0.98速度比与毫秒级触拍验收依裁判总数确认，未由录像测量。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频约 3.12 s，机器人右手持球拍，红色拍面击打橙色球；候选方式“击打”，现有接触方式词表无此项，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频约 3.12–3.36 s，机器人双脚接地支撑身体，右臂持拍颠球。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.72,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.60–3.80 s，第三次反弹后的球回落至球拍附近，随后继续落到拍面以下，没有第四次向上反弹；视频 3.84–4.16 s 球继续落向地面并低位弹跳。log step 93 的本局服务终局结果仅确认3次有效颠球，结束原因为球触地。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 3.60–3.80 s，机器人右手持拍调整并上抬，尝试继续击打下落球；球穿过拍面高度后继续下降，未见再次击起。候选方式“击打”，词表无此项，故按实际击打尝试用其他，不声称建立了球拍与球的第四次接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 3.60–4.16 s，机器人双脚仍接地支撑身体，在站立状态下持拍追球。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达5，第5次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达6，第6次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达7，第7次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达8，第8次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达9，第9次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达10，第10次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达11，第11次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达12，第12次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达13，第13次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达14，第14次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达15，第15次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达16，第16次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达17，第17次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达18，第18次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达19，第19次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.16,
+      "log_step": 93,
+      "text_zh": "观察：视频 3.72–4.16 s 球从球拍下方继续下落，未见后续有效反弹；log step 93 的 episode=6 服务结果为 stages_reached=3/20、done=true，球触地终止。推断：累计有效计数未达20，第20次颠球未完成，也未进入该次尝试。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": 3
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_4",
+   "t": 3.72,
+   "log_step": 93,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 3.60–3.80 s，机器人仍右手持拍追踪下落球，但球越过拍面高度后继续下降；视频 3.84–4.16 s 可见球落到脚边低处。log step 93 的机器人服务结果确认最后一局 episode=6 已完成3次有效颠球，并因球触地失败。推断：第4次击球未接续成功，随后未修复，故断点是第4次颠球。"
+  },
+  "unmet": {
+   "bounce_1": true,
+   "bounce_2": true,
+   "bounce_3": true,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 3.72,
+    "log_step": 93,
+    "evidence_zh": "观察：视频 3.60–3.80 s，橙球下落时机器人右手球拍没有在球经过拍面高度时将其击回，球由拍边附近落到拍下；视频 3.84–4.16 s 球继续下降，机器人仍抬拍。推断：机器人未跟上下落球完成下一次动态拦截。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "反馈控制",
+    "source_zh": "日志",
+    "evidence_zh": "推断：agent 的闭环击球控制在失去检测后仍依赖旧观测外推，未能恢复拦截。log step 62 的脚本只在检测到球时更新历史位置，并用历史拟合弹道；log step 72 将击球阶段的横向拍位更新冻结，log step 82 又加入预测像素偏差超过16即拒绝检测的门槛。该脚本经 log step 90–91 修改后运行最后一局；log step 93 的本局末两条调试输出均为 source=none、ball=null、phase=hit，仍生成拦截预测与拍姿，随后服务报告球触地。视频 3.60–4.16 s 同时显示球下落而球拍未将球击回，支持失去有效反馈后控制未纠正的归因。"
+   }
+  ],
+  "gap_zh": "没有逐帧状态或毫秒级接触日志，25 fps视频不能单独验证0.98速度比、0.02秒接触窗口；前三次有效性由本局裁判3次计数与三次可见反弹联合确认。日志只保留本局末两条跟踪调试输出，无法确定首次丢失检测的时刻，或区分丢失检测的具体视觉原因与弹道外推误差；未另标感知根因。所有节点时刻和断点均取自视频，未将agent工作时间或服务仿真时间换算成视频时间。"
+ },
+ "summary_zh": "受评的是最后一局 episode=6：机器人右手持拍完成前三次有效颠球，最终进度3/20；先前会话达到17次的测试不计入本局。视频约 3.72 s，第4次下落球未被球拍击回，随后球触地结束；日志显示控制器在检测缺失时仍外推并继续击球，支持反馈控制失效。缺少逐帧状态和首次丢失检测前后的完整调试记录，无法进一步确认跟踪丢失的具体原因或毫秒级验收细节。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "三次反弹时刻（视频约2.3/2.8/3.2 s）、断点bounce_4、动态时机现象与反馈控制根因均与视频和log step 93的服务读数（stages_reached=3、source=none、ball=null、球触地）一致，未发现需改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:59:51Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--nx7ih2g/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--nx7ih2g/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "7cfc198ee56d76494aa4e98d61e1d0c85dd51b9d5ec33d070310d8ed696d094f",
+   "video_sha256": "11cd73506422e75131fd3f514768fd4d985d476b2038b3e94591871f65044ccd",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--pehkp9o.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--pehkp9o.json
new file mode 100644
index 0000000000000000000000000000000000000000..6fe53cf50163c331c2eb50534941cc91803e7588
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--pehkp9o.json
@@ -0,0 +1,825 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard--pehkp9o",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "live_success": 1,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 1525,
+   "sim_steps": 1815,
+   "calls": 31,
+   "rows": 1805,
+   "resets": 1,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 6880.2,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.939,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.468,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard--pehkp9o/replay.mp4",
+  "duration_s": 15.28,
+  "fps": 25.0,
+  "frames": 382,
+  "sha256": "811380e21890b3605d3a3facf82cf841ad6a2ac9a4a14cff1f722a5e45b5aa16"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 20,
+  "final": 20,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.4,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.4,
+      "log_step": 113,
+      "text_zh": "观察：视频 2.24–2.56 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 2.40 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第1次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 2.4,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第1次有效颠球，达到时刻约为视频 2.40 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.24–2.56 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.24–2.56 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.04,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 130,
+      "text_zh": "观察：视频 2.80–3.20 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 3.04 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第2次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 3.04,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第2次有效颠球，达到时刻约为视频 3.04 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.80–3.20 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.80–3.20 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.68,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 142,
+      "text_zh": "观察：视频 3.44–3.84 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 3.68 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第3次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 3.68,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第3次有效颠球，达到时刻约为视频 3.68 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 3.44–3.84 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 3.44–3.84 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.24,
+      "log_step": 145,
+      "text_zh": "观察：视频 4.08–4.40 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 4.24 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第4次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 4.24,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第4次有效颠球，达到时刻约为视频 4.24 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 4.08–4.40 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 4.08–4.40 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.44,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.44,
+      "log_step": 151,
+      "text_zh": "观察：视频 5.28–5.60 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 5.44 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第6次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 5.44,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第5次有效颠球，达到时刻约为视频 5.44 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 5.28–5.60 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。 本步还包括视频 4.72–5.04 s 的一次物理反弹；其未计数是依据日志拟合与终局计数作出的推断。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 5.28–5.60 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.08,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.08,
+      "log_step": 162,
+      "text_zh": "观察：视频 5.92–6.24 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 6.08 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第7次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 6.08,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第6次有效颠球，达到时刻约为视频 6.08 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 5.92–6.24 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 5.92–6.24 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.72,
+      "log_step": 169,
+      "text_zh": "观察：视频 6.48–6.88 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 6.72 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第8次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 6.72,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第7次有效颠球，达到时刻约为视频 6.72 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 6.48–6.88 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 6.48–6.88 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.36,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.36,
+      "log_step": 172,
+      "text_zh": "观察：视频 7.12–7.52 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 7.36 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第9次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 7.36,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第8次有效颠球，达到时刻约为视频 7.36 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 7.12–7.52 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 7.12–7.52 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.48,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.48,
+      "log_step": 178,
+      "text_zh": "观察：视频 8.24–8.64 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 8.48 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第11次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 8.48,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第9次有效颠球，达到时刻约为视频 8.48 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 8.24–8.64 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。 本步还包括视频 7.76–8.08 s 的一次物理反弹；其未计数是依据日志拟合与终局计数作出的推断。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 8.24–8.64 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.12,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.12,
+      "log_step": 181,
+      "text_zh": "观察：视频 8.88–9.20 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 9.12 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第12次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 9.12,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第10次有效颠球，达到时刻约为视频 9.12 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 8.88–9.20 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 8.88–9.20 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.68,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.68,
+      "log_step": 184,
+      "text_zh": "观察：视频 9.44–9.76 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 9.68 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第13次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 9.68,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第11次有效颠球，达到时刻约为视频 9.68 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 9.44–9.76 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 9.44–9.76 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.24,
+      "log_step": 209,
+      "text_zh": "观察：视频 10.00–10.40 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 10.24 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第14次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 10.24,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第12次有效颠球，达到时刻约为视频 10.24 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 10.00–10.40 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 10.00–10.40 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.8,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.8,
+      "log_step": 214,
+      "text_zh": "观察：视频 10.64–10.96 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 10.80 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第15次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 10.8,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第13次有效颠球，达到时刻约为视频 10.80 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 10.64–10.96 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 10.64–10.96 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [
+     [
+      11.44,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 11.44,
+      "log_step": 220,
+      "text_zh": "观察：视频 11.28–11.60 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 11.44 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第16次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 11.44,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第14次有效颠球，达到时刻约为视频 11.44 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 11.28–11.60 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 11.28–11.60 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [
+     [
+      12.08,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 12.08,
+      "log_step": 222,
+      "text_zh": "观察：视频 11.92–12.24 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 12.08 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第17次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 12.08,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第15次有效颠球，达到时刻约为视频 12.08 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 11.92–12.24 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 11.92–12.24 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [
+     [
+      12.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 12.72,
+      "log_step": 224,
+      "text_zh": "观察：视频 12.48–12.88 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 12.72 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第18次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 12.72,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第16次有效颠球，达到时刻约为视频 12.72 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 12.48–12.88 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 12.48–12.88 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [
+     [
+      13.36,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 13.36,
+      "log_step": 226,
+      "text_zh": "观察：视频 13.20–13.52 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 13.36 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第19次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 13.36,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第17次有效颠球，达到时刻约为视频 13.36 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 13.20–13.52 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 13.20–13.52 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [
+     [
+      14.0,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 14.0,
+      "log_step": 228,
+      "text_zh": "观察：视频 13.84–14.16 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 14.00 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第20次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 14.0,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第18次有效颠球，达到时刻约为视频 14.00 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 13.84–14.16 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 13.84–14.16 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [
+     [
+      14.64,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 14.64,
+      "log_step": 230,
+      "text_zh": "观察：视频 14.48–14.80 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 14.64 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第21次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 14.64,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第19次有效颠球，达到时刻约为视频 14.64 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 14.48–14.80 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 14.48–14.80 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [
+     [
+      15.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 15.24,
+      "log_step": 232,
+      "text_zh": "观察：视频 15.12–15.24 s，机器人右手握持红色球拍，橙色球降至拍面后向上反弹；视频 15.24 s 球已回到开局释放球高度附近或更高。该段为从放球后顺数的第22次物理反弹，双脚持续着地支撑。"
+     },
+     {
+      "t": 15.24,
+      "log_step": 232,
+      "text_zh": "推断：此段对应第20次有效颠球，达到时刻约为视频 15.24 s（目视锚点，约±0.2 s）。服务裁判在 log step 232 报告最后一局 outcome=success、stages=20，log step 233 确认 episode=2；因此全部20个有效计数最终成立。物理反弹与有效计数的对应结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述推定：第5、10次物理反弹不计数，其余按顺序计入；没有逐次原生计数日志，不能仅凭画面验证0.98速度比及0.02秒接触门槛。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 15.12–15.24 s，机器人右手持拍，以红色拍面击起橙色球。候选方式为“击打”，现有接触方式无对应值，故用其他；持拍柄不重复记录。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 15.12–15.24 s 及全景连续画面，机器人双脚着地，支撑站立颠球。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": null,
+ "summary_zh": "受评录像对应重置后的第2局：机器人右手持拍、双脚站立，完成全部20个有效颠球节点；服务裁判在 log step 232 返回 success、stages=20，log step 233 确认终局，视频 15.24 s 仍持拍且球在上升。视频可见22次物理反弹，结合 log step 150、177 的 agent 轨迹拟合及 log step 197、235 的 agent 自述和最终20阶段结果，推断第5、10次物理反弹未计入有效数；逐节点时刻以检查过的上升帧作近似锚点，并非逐次裁判锁存时刻。无状态日志或逐次原生计数，25 fps录像无法独立验证速度比和毫秒接触门槛；三项原生指标未直接报告数值，保留null，20个节点的最终验收依据服务裁判成功结果。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "20个节点均为成功，依据服务裁判 log step 232 的 success/stages=20（graded episode 2，control_steps 读数），视频可见机器人持拍站立颠球至 15.24 s；各节点时刻为目视近似锚点并已标注为推断，未发现节点值、接触或时钟错误；无失败，故无断点/现象。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:50:08Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--pehkp9o/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--pehkp9o/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "acb33d89b377ab397ba3af9d112e7ada9f43f52d89b1f117b046b1942e35da53",
+   "video_sha256": "811380e21890b3605d3a3facf82cf841ad6a2ac9a4a14cff1f722a5e45b5aa16",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--yej6xon.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--yej6xon.json
new file mode 100644
index 0000000000000000000000000000000000000000..27063d86839a6d2e7f4ffef680fc67a7ec266de0
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard--yej6xon.json
@@ -0,0 +1,558 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard--yej6xon",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 0,
+   "sim_steps": 0,
+   "calls": 1,
+   "rows": 0,
+   "resets": 0,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 6883.5,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 4,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.986,
+   "peak_processes": 11,
+   "peak_disk_gb": 0.381,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard--yej6xon/replay.mp4",
+  "duration_s": 0.04,
+  "fps": 25.0,
+  "frames": 1,
+  "sha256": "7b6f2ee30688fd30534b91a716fae98d15aff82479297859f442c76f50cfd16f"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第1次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第2次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第3次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第4次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第5次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第6次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第7次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第8次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第9次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第10次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第11次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第12次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第13次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第14次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第15次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第16次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第17次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第18次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第19次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 8,
+      "text_zh": "观察：受评视频唯一帧（视频 0.00 s，亦为末帧）中，橙色球仍在释放杯处，机器人双脚站立，红色球拍垂在右手旁，未发生击球；log step 8 的服务初始观测为 control_steps=0，评分器最终字段 n_actions=0、sim_steps=0、rows=0。推断：受评回合没有推进，第20次有效颠球未达成；日志中的离线测试不计入该节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分器记录 n_actions=0、sim_steps=0、rows=0；没有向实时受评回合交付控制轨迹，未开展本节点的颠球接触。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_1",
+   "t": 0.0,
+   "log_step": 8,
+   "where": "首步前",
+   "evidence_zh": "观察：视频 0.00 s 是全片唯一帧，球仍在释放杯处、球拍垂于右手旁；log step 8 仅调用服务 get_obs，返回 control_steps=0。评分器最终记录 n_actions=0、sim_steps=0、rows=0、resets=0；log step 38 创建实时驱动脚本，log step 39 声明先做离线假物理测试，直到 log step 123 仍在运行自行构造的离线测试。推断：第1次颠球的受评控制从未开始，断点定位首步前。"
+  },
+  "unmet": {
+   "bounce_1": false,
+   "bounce_2": false,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  },
+  "phenomena": [],
+  "root_causes": [],
+  "gap_zh": "观察：输入视频存在但仅有视频 0.00 s 的初始帧，全长视频 0.04 s；没有状态日志，评分器未报告 bounces、ball_height_m、grip_break_max_s 原生指标。log step 39—123 的触拍、落地及计数来自 agent 自建离线假物理测试，并非受评服务；log step 123 后无最终完成报告。推断：本次属于未交付控制轨迹的 agent 侧失败，依 D23 将现象与根因均留空；缺少实际受评控制过程，不能把离线测试中的控制缺陷直接归为本次物理失败，也不能仅从零动作反推规划或交付操作错误。评分器 infra_error=0、grader_error=0、sim_error=0，日志未显示外部服务中断。"
+ },
+ "summary_zh": "观察：受评视频仅有视频 0.00 s 的初始画面，机器人持拍站立、球仍在释放杯处；评分器记录零控制动作、零仿真步，20个颠球节点全部未达成。推断：失败停在第1次颠球的首步前；agent 创建了实时驱动脚本，但直到 log step 123 仍在调试离线假物理测试，未交付受评控制轨迹。观察：没有实际受评运动或状态日志可支持物理失败现象及根因，依 D23 留空，原生指标未报告。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "20个节点全部未达成、断点 bounce_1 首步前、现象与根因留空并附 gap_zh，均符合 D23(a)；评分器 n_actions=0/sim_steps=0 与单帧视频一致，未发现需改的内容。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:01:50Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--yej6xon/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard--yej6xon/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "1e3e77f5d545c38a59a4b3edd925085e9366b33d90a50e1c215086837b18f2aa",
+   "video_sha256": "7b6f2ee30688fd30534b91a716fae98d15aff82479297859f442c76f50cfd16f",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--8fksne7.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--8fksne7.json
new file mode 100644
index 0000000000000000000000000000000000000000..46213393aa7841d70b879e6d65d233ea5ff7def2
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--8fksne7.json
@@ -0,0 +1,836 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard-realtime--8fksne7",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "live_success": 1,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 1143,
+   "sim_steps": 2466,
+   "calls": 1200,
+   "rows": 2568,
+   "resets": 4,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 1581.5,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "realtime": 1,
+   "slowdown": 1.0,
+   "clock_s": 24.665,
+   "max_lag_s": 0.0437,
+   "stalls": 0,
+   "stalled_s": 0.0,
+   "stalled_share": 0.0,
+   "held_steps": 430,
+   "late_steps": 0,
+   "requests_refused": 4,
+   "request_busy_s": 6.335,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--8fksne7/replay.mp4",
+  "duration_s": 11.44,
+  "fps": 25.0,
+  "frames": 286,
+  "sha256": "03f03bc7d638f88133cd5e1b0126a65710fb31eea4a9e7ed7ad36fca5227bdba"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 20,
+  "final": 20,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.6,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.36,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 2.20–2.36 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 2.68–2.84 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 2.6,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第1次有效颠球约在视频 2.36–2.84 s 的两轮上升之间完成，intervals起点取视频 2.60 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.20–2.84 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.08,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 2.68–2.84 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 3.16–3.32 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 3.08,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第2次有效颠球约在视频 2.84–3.32 s 的两轮上升之间完成，intervals起点取视频 3.08 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.68–3.32 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.56,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.32,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 3.16–3.32 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 3.68–3.80 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 3.56,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第3次有效颠球约在视频 3.32–3.80 s 的两轮上升之间完成，intervals起点取视频 3.56 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 3.16–3.80 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.04,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.8,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 3.68–3.80 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 4.12–4.28 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 4.04,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第4次有效颠球约在视频 3.80–4.28 s 的两轮上升之间完成，intervals起点取视频 4.04 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 3.68–4.28 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.52,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.28,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 4.12–4.28 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 4.64–4.76 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 4.52,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第5次有效颠球约在视频 4.28–4.76 s 的两轮上升之间完成，intervals起点取视频 4.52 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 4.12–4.76 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.0,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 4.76,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 4.64–4.76 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 5.12–5.24 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 5.0,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第6次有效颠球约在视频 4.76–5.24 s 的两轮上升之间完成，intervals起点取视频 5.00 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 4.64–5.24 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.48,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.24,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 5.12–5.24 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 5.56–5.72 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 5.48,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第7次有效颠球约在视频 5.24–5.72 s 的两轮上升之间完成，intervals起点取视频 5.48 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 5.12–5.72 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [
+     [
+      5.94,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 5.72,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 5.56–5.72 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 6.00–6.16 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 5.94,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第8次有效颠球约在视频 5.72–6.16 s 的两轮上升之间完成，intervals起点取视频 5.94 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 5.56–6.16 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.38,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.16,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 6.00–6.16 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 6.48–6.60 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 6.38,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第9次有效颠球约在视频 6.16–6.60 s 的两轮上升之间完成，intervals起点取视频 6.38 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 6.00–6.60 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [
+     [
+      6.84,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 6.6,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 6.48–6.60 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 6.92–7.08 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 6.84,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第10次有效颠球约在视频 6.60–7.08 s 的两轮上升之间完成，intervals起点取视频 6.84 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 6.48–7.08 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.3,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.08,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 6.92–7.08 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 7.36–7.52 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 7.3,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第11次有效颠球约在视频 7.08–7.52 s 的两轮上升之间完成，intervals起点取视频 7.30 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 6.92–7.52 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [
+     [
+      7.72,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.52,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 7.36–7.52 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 7.80–7.92 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 7.72,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第12次有效颠球约在视频 7.52–7.92 s 的两轮上升之间完成，intervals起点取视频 7.72 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 7.36–7.92 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.16,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 7.92,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 7.80–7.92 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 8.24–8.40 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 8.16,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第13次有效颠球约在视频 7.92–8.40 s 的两轮上升之间完成，intervals起点取视频 8.16 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 7.80–8.40 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [
+     [
+      8.62,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.4,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 8.24–8.40 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 8.68–8.84 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 8.62,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第14次有效颠球约在视频 8.40–8.84 s 的两轮上升之间完成，intervals起点取视频 8.62 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 8.24–8.84 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.06,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 8.84,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 8.68–8.84 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 9.12–9.28 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 9.06,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第15次有效颠球约在视频 8.84–9.28 s 的两轮上升之间完成，intervals起点取视频 9.06 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 8.68–9.28 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.5,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.28,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 9.12–9.28 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 9.56–9.72 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 9.5,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第16次有效颠球约在视频 9.28–9.72 s 的两轮上升之间完成，intervals起点取视频 9.50 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 9.12–9.72 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.94,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 9.72,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 9.56–9.72 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 10.00–10.16 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 9.94,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第17次有效颠球约在视频 9.72–10.16 s 的两轮上升之间完成，intervals起点取视频 9.94 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 9.56–10.16 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.38,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.16,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 10.00–10.16 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 10.44–10.60 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 10.38,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第18次有效颠球约在视频 10.16–10.60 s 的两轮上升之间完成，intervals起点取视频 10.38 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 10.00–10.60 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [
+     [
+      10.82,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 10.6,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频 10.44–10.60 s，机器人右手红色拍面下方承接落球后，橙球反弹上升；下一轮相同接触和上升见视频 10.88–11.04 s。机器人双脚支撑地面，球拍随右手运动。"
+     },
+     {
+      "t": 10.82,
+      "log_step": 61,
+      "text_zh": "推断：log step 61 中服务裁判（机器人API返回的原生回合状态）确认第5局 stages_reached=20/20、reason=success；视频共见21轮物理反弹，无法确定其中哪一轮未计入有效次数。因此第19次有效颠球约在视频 10.60–11.04 s 的两轮上升之间完成，intervals起点取视频 10.82 s 为导航估计，误差约±0.3 s；不是将脚本REBOUND序号等同裁判阶段。此event达到后锁存至结尾。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 10.44–11.04 s 附近，机器人右手所持红色球拍拍面击打橙球，球随即向上离拍；有效计数对应的具体一轮存在约一轮歧义。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [
+     [
+      11.4,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 11.4,
+      "log_step": null,
+      "text_zh": "观察：视频 11.28–11.40 s，橙球在红色拍面附近由下降转为上升；视频 11.40 s 末帧右手仍持拍、双脚仍落地支撑。"
+     },
+     {
+      "t": 11.4,
+      "log_step": 61,
+      "text_zh": "推断：log step 57 启动的第5局是最后受评回合，log step 61 服务裁判返回 control_steps=1143、sim_time_s=11.43、stages_reached=20、stages_total=20、reason=success；任务控制步长0.01 s，1143×0.01=11.43 s与视频时长11.44 s吻合，且已核对视频约2.00 s的释放器放球事件，因此第20次有效颠球在视频约11.4 s完成。该时间来自受评回合控制步计数与末段画面，不来自agent会话时间；评分器reward.live_success=1支持全部目标完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 11.28–11.40 s，机器人右手所持红色球拍拍面击打橙球，球向上离拍。候选方式“击打”，现有接触方式词表无击打项，依S16标其他；持拍柄接触不重复标注。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：完整视频 0.00–11.40 s 中机器人以双脚支撑地面，原地持拍颠球；未见跌倒或以手、膝、躯干撑地。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": null,
+ "summary_zh": "最后受评的第5局中，机器人站立、右手持拍，将释放器落下的球连续击起；服务裁判在log step 61确认20/20阶段成功，全部20个颠球节点最终为true，无失败断点。逐帧检查视频 0.00–11.40 s（含末秒）并对照成功参考录像后，可见21轮物理反弹；缺少逐次原生计数及毫秒接触/速度记录，无法确定其中哪一轮未计数，前19个节点首次时间用相邻两轮上升时刻的中点估计（约±0.3 s），第20个节点由末段画面与回合终态定位至视频约11.4 s。评分器reward未报告DAG所列bounces、ball_height_m和grip_break_max_s原生字段，三个指标均保留null；脚本的REBOUND计数不充当裁判有效计数。",
+ "conf": 6,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "成功运行，20/20 节点 final=true，与服务裁判 stages_reached=20 相符；时间用控制步 1143×0.01 与视频时长对应，时钟标注清楚，无失败断点，接触标注合理。未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:04:48Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--8fksne7/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--8fksne7/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "8e9f590f20e3ae60653ed083e06b6a5eef71ed36019cf2593871fbf16403b8d8",
+   "video_sha256": "03f03bc7d638f88133cd5e1b0126a65710fb31eea4a9e7ed7ad36fca5227bdba",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--98hy2z3.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--98hy2z3.json
new file mode 100644
index 0000000000000000000000000000000000000000..e6c69fc42de6ab0ef6abe5a8d1eed8d6fc6c14a3
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--98hy2z3.json
@@ -0,0 +1,466 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard-realtime--98hy2z3",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 259,
+   "sim_steps": 1455,
+   "calls": 19,
+   "rows": 6000,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 5948.4,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "realtime": 1,
+   "slowdown": 1.0,
+   "clock_s": 14.589,
+   "max_lag_s": 0.0147,
+   "stalls": 0,
+   "stalled_s": 0.0,
+   "stalled_share": 0.0,
+   "held_steps": 0,
+   "late_steps": 0,
+   "requests_refused": 0,
+   "request_busy_s": 2.171,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--98hy2z3/replay.mp4",
+  "duration_s": 2.6,
+  "fps": 25.0,
+  "frames": 65,
+  "sha256": "4508e322789a7b4f7b3015559805c333d767d818f632b5ab4621d13f48984fc4"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.84,
+      "log_step": null,
+      "text_zh": "观察：视频0.40–1.84 s，机器人右手持红色球拍抬到身前，随后下降迎球，拍面仍近竖直。"
+     },
+     {
+      "t": 2.28,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s，橙球从释放杯下落，经过拍缘附近后继续降到腿旁和地面，没有向上返回释放高度；第1次有效颠球未完成。"
+     },
+     {
+      "t": null,
+      "log_step": 145,
+      "text_zh": "观察：log step 145 的机器人服务返回episode=6、control_steps=259、done=True、reason=failed，与本视频单局约2.6 s的结束相符；本次标注仅针对该最后一局。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "推断：视频2.20 s橙球到达右手所持红拍上缘附近，视频2.24 s被拍面遮挡，视频2.28 s从拍面左侧出现，随后明显向画面左下方偏出，支持拍面与球发生擦边击打，但没有有效向上回弹；候选接触方式“击打”，现有词表无对应项，故用其他。接触瞬间及具体接触点不能由25 fps画面精确确认。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.00–2.56 s，机器人双脚始终在地面支撑身体，在站立状态下右臂持拍迎球。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第2次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第3次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第4次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第5次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第6次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第7次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第8次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第9次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第10次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第11次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第12次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第13次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第14次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第15次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第16次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第17次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第18次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第19次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.56,
+      "log_step": null,
+      "text_zh": "观察：视频2.00–2.56 s只出现首次放球下落，橙球未被击回释放高度即落地，未出现连续颠球。推断：第1次有效颠球未完成，依累计计数及前置顺序，第20次有效颠球也未达到；机器人尚未单独尝试此计数。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_1",
+   "t": 2.2,
+   "log_step": 145,
+   "where": "操作中",
+   "evidence_zh": "观察：视频2.16–2.36 s，机器人右手将近竖直拍面置于下落球附近，橙球经过拍缘后继续降至腿旁，视频2.52 s已到地面附近，没有形成首次达标上升；log step 145的最后一局服务状态为failed。失败发生在第1次颠球操作中。"
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 2.2,
+    "log_step": null,
+    "evidence_zh": "观察：视频1.80–2.32 s，机器人迎球时红拍面近竖直，橙球经过拍缘附近后向左下方偏出，并未向上弹回。推断：迎球拍面的朝向未达到向上颠球所需姿态，首次颠球失败；这里可见的是拍面朝向错误，没有充分证据另判击球过早或过晚。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "观察：log step 125的控制脚本以离线正运动学计算水平拍面（法向-z），按预先生成的关节序列迎球；log step 130的修改仍用该模型计算拍面，log step 136加入两段抬臂轨迹，log step 144恢复grip=1并减慢下降段，log step 145沿用此脚本执行最后一局。视频1.80–2.32 s显示机器人手臂已到身前，实际拍面仍近竖直。推断：该持拍与关节控制未能把模型中的水平拍面实现到实际接触中，预设轨迹也没有根据实际拍面朝向修正，导致迎球姿态失败；现有证据不足以进一步区分握持传力、模型误差与跟踪控制。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志，也没有最后一局的原生有效颠球数、球高度或持拍中断时长字段，故三项native_metrics均为null。视频为25 fps，球在视频2.24 s附近受拍面遮挡，擦边击打由前后运动推断，无法确定接触点、接触力和毫秒级时长；这些不影响球从未达标上升的判断。agent在log step 145、147将失败归为“无摩擦转轴”并声称实例不可完成，但缺少独立接触力矩证据，不能据此认定任务或环境根因；成功参考只用于对照水平拍面和击后上升的可见线索。",
+  "unmet": {
+   "bounce_1": false,
+   "bounce_2": false,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  }
+ },
+ "summary_zh": "受评的是最后第6局：机器人右手持拍抬臂并迎接下落球，但视频2.20–2.36 s拍面近竖直，球经过拍缘后向下偏出，随后落地，20个有效颠球节点均未完成。断点为第1次颠球，现象是迎球拍面位姿不准，控制脚本的水平拍面目标未在实际执行中实现；无状态日志及接触力矩记录，擦边接触与具体控制机理仍有不确定性。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "视频2.6 s内球从释放杯落下、未被拍面击回，bounce_1至20均未达成，断点bounce_1、位姿不准与控制根因与帧相符，未发现需要改动的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:04:41Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--98hy2z3/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--98hy2z3/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "f50e2c29a025ccf114c41fbd5e9f584a5d79cbcb5085259ff095404083974787",
+   "video_sha256": "4508e322789a7b4f7b3015559805c333d767d818f632b5ab4621d13f48984fc4",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--cek76dv.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--cek76dv.json
new file mode 100644
index 0000000000000000000000000000000000000000..1ab3ccedfb639c515d520c99d9b0ec22424d6b2a
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--cek76dv.json
@@ -0,0 +1,488 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard-realtime--cek76dv",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 252,
+   "sim_steps": 1351,
+   "calls": 466,
+   "rows": 1795,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2957.6,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "realtime": 1,
+   "slowdown": 1.0,
+   "clock_s": 13.74,
+   "max_lag_s": 0.1527,
+   "stalls": 0,
+   "stalled_s": 0.0,
+   "stalled_share": 0.0,
+   "held_steps": 267,
+   "late_steps": 0,
+   "requests_refused": 2,
+   "request_busy_s": 2.758,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--cek76dv/replay.mp4",
+  "duration_s": 2.56,
+  "fps": 25.0,
+  "frames": 64,
+  "sha256": "ba65f7915a4ebb477cecbb284d2b2abc5ca5a4f1c5e3f501e210dec1a9bd7b9a"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.64,
+      "log_step": 209,
+      "text_zh": "观察：视频 0.48–0.80 s，机器人右手将红色球拍从大腿旁的竖直姿态转成斜向外伸，开始为首次颠球调拍；log step 209 启动最后一次实时尝试。"
+     },
+     {
+      "t": 2.32,
+      "log_step": 210,
+      "text_zh": "观察：视频 2.00–2.52 s，橙色球离开释放器后持续下落，经过右手球拍前方并落地；红色拍面仍在髋部旁，未与球相遇，也没有击球后的上升。log step 210 的最后一局结束输出为“the ball touched the floor”，阶段为 0/20。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–2.52 s，机器人双足接触地面支撑身体，同时右手持拍调整位置；未观察到拍面接触小球，不记实际击打接触。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第2次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第3次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第4次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第5次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第6次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第7次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第8次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第9次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第10次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第11次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第12次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第13次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第14次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第15次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第16次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第17次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第18次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第19次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": 210,
+      "text_zh": "推断：视频 0.00–2.52 s 仅见一次放球，随后小球未被球拍击起就落地，未出现任何连续颠球；log step 210 最后一局结束输出阶段 0/20。因此累计第20次有效颠球未达到，且本局未进入该次尝试。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_1",
+   "t": 2.32,
+   "log_step": 210,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 0.48–0.80 s，机器人已调整右手球拍，为首次颠球作准备；视频 2.16–2.40 s，小球下落穿过拍面所在高度，球拍仍偏在身体右侧，未截住小球。视频 2.52 s 小球已到地面，log step 210 最后一局输出“the ball touched the floor”，阶段 0/20；首次颠球未完成。"
+  },
+  "unmet": {
+   "bounce_1": false,
+   "bounce_2": false,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 2.32,
+    "log_step": 210,
+    "evidence_zh": "观察：视频 2.16–2.40 s，机器人右手球拍未跟上下落小球，拍面留在髋部旁，小球从拍面前方降至膝部以下；视频 2.52 s 小球落地。本次漏接发生在运动中的球上，未出现击球反弹。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 0.8,
+    "log_step": 212,
+    "source_zh": "日志",
+    "evidence_zh": "推断：log step 210 最后一局调平计划的右腕偏航目标为 +1.03 rad，而 log step 212 从该局保存的关节反馈读得右腕偏航约 -0.55 rad、肩俯仰约 +0.64 rad；其模型原先预期肩俯仰约 +0.38 rad。视频 0.80–2.40 s 的球拍持续停在髋部旁，未实现计划中的向上承球姿态，说明该控制器未获得并纠正所需的关节与拍面跟踪。现有证据不足以将偏差归为某一种力矩或摩擦机制。"
+   },
+   {
+    "cat": "认知",
+    "item": "感知",
+    "t": 1.76,
+    "log_step": 227,
+    "source_zh": "日志",
+    "evidence_zh": "推断：log step 227 展示的受评控制器代码把拍面像素射线与固定高度 self.h0 相交，并用所得位置与关节正运动学之差估计底座位移；log step 211 读得拍面估计坐标约 (0.617, -0.988) m，log step 210 因而推得底座位移 (0.683, -0.781) m、重规划误差 0.4151 m。视频 0.80–1.76 s 机器人仍在原处，球拍位于身体旁；log step 212 用同局关节反馈估计的拍面高度约 0.646 m，也不同于所用固定高度 0.785 m。错误位置估计使纠偏重规划未被接受；此归因依据代码与数据差异，而非只采信 agent 的解释。"
+   }
+  ],
+  "gap_zh": "没有状态日志或毫秒级原生接触记录，无法独立确定电机、握持摩擦对姿态偏差各自的贡献；log step 213–219 对力矩和摩擦锁止的解释属于 agent 的事后假设，未据此标成任务不可行。reward 未报告 bounces、ball_height_m、grip_break_max_s 三项原生指标，故其 value 保留 null；节点未达由视频及最后一局阶段输出共同确认。log step 228 之后的离线修改与模拟客户端测试未重新执行于受评局。"
+ },
+ "summary_zh": "受评的最后一局中，机器人调整右手球拍但拍面始终留在髋部旁；视频 2.00–2.52 s 小球释放后未经击打便落地，20 个颠球节点均未达到，断点为第1次颠球。日志支持关节跟踪未达预期和拍面位置估计错误共同导致首次漏接；其后的控制器修订仅作离线测试。没有状态日志，无法验证 agent 所称力矩不足或摩擦锁止的具体机制，不能据此判定任务不可行。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 7,
+    "summary_zh": "节点、断点、时钟与受评局（最后一局，log step 209–210）均与视频一致，20 个节点全未达到、断点 bounce_1 成立。问题在现象：球拍从未移到释放点下方，球落点是静态已知位置，球拍也没有任何向球的动作，应记位姿不准而非动态时机；根因（控制+感知）有代码与日志依据，保留。",
+    "n_issues": 1,
+    "n_major": 1
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点（20个均未达到）、断点 bounce_1、时钟、受评局（最后一局）与视频一致；第1轮的现象分歧已有规则依据（正在下落的球漏接用动态时机），无新证据，根因有代码与日志支撑。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:05:34Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--cek76dv/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--cek76dv/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--cek76dv/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--cek76dv/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "7abdaaeeee8b3f611433f268e6e96f768ce76fd3c64f099068bcc813295a4ae2",
+   "video_sha256": "ba65f7915a4ebb477cecbb284d2b2abc5ca5a4f1c5e3f501e210dec1a9bd7b9a",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pndceqs.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pndceqs.json
new file mode 100644
index 0000000000000000000000000000000000000000..61e4fb011486098440d8dfbec15eeb8eefe3c6bb
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pndceqs.json
@@ -0,0 +1,463 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pndceqs",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 273,
+   "sim_steps": 1657,
+   "calls": 1400,
+   "rows": 1788,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2306.7,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "realtime": 1,
+   "slowdown": 1.0,
+   "clock_s": 17.549,
+   "max_lag_s": 0.2781,
+   "stalls": 0,
+   "stalled_s": 0.0,
+   "stalled_share": 0.0,
+   "held_steps": 516,
+   "late_steps": 632,
+   "requests_refused": 6,
+   "request_busy_s": 3.977,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pndceqs/replay.mp4",
+  "duration_s": 2.76,
+  "fps": 25.0,
+  "frames": 69,
+  "sha256": "bc05d4f3114f40ebad9112be2a285fef9ab059d5bceeabfcfee43023219a6993"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.24,
+      "log_step": null,
+      "text_zh": "观察：视频 0.0–2.0 s，机器人右手持红色球拍从身侧抬至落球下方；视频 2.20–2.28 s，橙球靠近拍面后由下降转为上升。视频 2.32–2.60 s，球向躯干一侧偏移后下降，视频 2.68–2.72 s 已落到拍面下方，未形成连续有效颠球。"
+     },
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 213 启动最后一次重置后的 attempt 6，resets_left=0；log step 215 的实时机器人服务最终结果为 stages_reached=0、stages_total=20、control_steps=273，outcome 为球碰到拍面以外的物体。该服务结果与视频末尾球从握拍手一侧落下相符，第1次有效颠球未被裁判锁存；agent 脚本自行打印的 impact 和 apex 不等于裁判计数。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.20–2.28 s，机器人右手所持红色球拍与橙球发生击打，球由下降转为上升；候选接触方式“击打”，依S16使用其他。末尾裁判报告的拍面外接触无法从此视角精确区分是手、拍柄还是其他身体部位。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.0–2.72 s，机器人双脚支撑于地面，站立完成抬拍与击球尝试。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第2次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第3次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第4次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第5次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第6次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第7次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第8次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第9次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第10次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第11次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第12次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第13次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第14次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第15次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第16次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第17次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第18次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第19次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.72,
+      "log_step": 215,
+      "text_zh": "观察：log step 215 的最后一局服务结果为 stages_reached=0/20；视频 2.20–2.72 s 只见一次反弹后球偏向躯干一侧并落到拍面下方，未继续颠球。推断：第1次有效颠球尚未锁存，按顺序计数的第20次有效颠球也未达到，未单独开展该步。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "value": 0,
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better"
+  },
+  {
+   "name": "ball_height_m",
+   "value": null,
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better"
+  },
+  {
+   "name": "grip_break_max_s",
+   "value": null,
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_1",
+   "t": 2.24,
+   "log_step": 215,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 2.20–2.28 s，右手所持球拍使落球反弹，但视频 2.32–2.60 s 球路逐渐偏向躯干一侧，随后从握拍手附近落到拍面下方；机器人未完成第1次有效颠球。log step 215 的最终服务结果确认有效阶段为0/20，并以球触及拍面以外物体判负。断点锚在首次击球后球路偏移开始处，未把终止规则另作步骤。"
+  },
+  "unmet": {
+   "bounce_1": false,
+   "bounce_2": false,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 2.24,
+    "log_step": 215,
+    "evidence_zh": "观察：视频 2.20–2.28 s，机器人持拍击球后橙球上升；视频 2.32–2.60 s，球相对拍面逐渐移向躯干一侧，下降时未回到红色拍面中央，视频 2.68–2.72 s 已在拍面下方。推断：此次击打的出球方向偏离可持续颠球所需的回拍球路；log step 215 确认球最终发生拍面外接触，且有效计数仍为0。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 2.24,
+    "log_step": 213,
+    "source_zh": "视频 / 日志",
+    "evidence_zh": "观察：log step 209、213 显示 agent 使用预计算关节表、腕部预旋转补偿与拍高往复控制，最后一局设置 OFF_Y=0.36、OFF_P=-0.22；视频 2.20–2.60 s 显示球被拍起后向机器人躯干一侧偏移，持拍手未将拍面重新置于下降球路下。推断：最终控制未实现所需的拍面姿态与回球跟踪，归为控制。日志中早先试验的拍柄滑转诊断不能证明最后一局的具体滑转量，故不细分控制子类。"
+   }
+  ],
+  "gap_zh": "没有逐帧状态或毫秒级接触记录；视频 2.60–2.72 s 的手、拍柄与身体在同一投影附近，最终非法接触的具体对象无法可靠区分。log step 215 只报告拍面外触碰及0/20阶段，没有逐次出入球速率或触拍时长，不能进一步判定首个反弹具体未满足哪项精确阈值。log step 213 的 impact/apex 为 agent 自行估计，不能用其“1.301”替代裁判有效颠球数；原生当前球高及最大持拍接触中断时长未报告，指标留空。"
+ },
+ "summary_zh": "机器人在最后一局抬起右手所持球拍并使落球反弹，但球路向躯干一侧偏移，下降时从握拍手附近落至拍面下方；log step 215 的裁判阶段为0/20，全部20个节点均未完成。断点为第1次颠球，现象为击打球路位姿不准，根因归为控制；视频仅见一次物理反弹，不能按 agent 脚本的 impact 计数记为有效颠球。缺少毫秒级接触与速度记录，无法确认具体非法触碰部位及最终握拍滑转量。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频仅 2.76 s（最后一局），球在约 2.0 s 落下，2.20–2.28 s 被拍面击起后沿躯干前方上升再落向握拍手/腰侧；20 个节点均未达成，服务结果 0/20，标注与帧和 log step 215 一致。断点 bounce_1、位姿不准、控制根因有证据，agent 自述的拍柄滑转已被标注者正确排除在观察之外。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:03:47Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pndceqs/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pndceqs/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "97311eabb0159cae3a1da7bc846d334362db29fcc9af68213940b4c5e4c21078",
+   "video_sha256": "bc05d4f3114f40ebad9112be2a285fef9ab059d5bceeabfcfee43023219a6993",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pwahmzu.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pwahmzu.json
new file mode 100644
index 0000000000000000000000000000000000000000..86f1bac9be0217f2839cfcaa87f04916f947c5a5
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pwahmzu.json
@@ -0,0 +1,669 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pwahmzu",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 0,
+   "sim_steps": 0,
+   "calls": 16,
+   "rows": 0,
+   "resets": 0,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 4835.8,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 4,
+   "realtime": 1,
+   "slowdown": 1.0,
+   "clock_s": 0.0,
+   "max_lag_s": 0.0,
+   "stalls": 0,
+   "stalled_s": 0.0,
+   "stalled_share": 0.0,
+   "held_steps": 0,
+   "late_steps": 0,
+   "requests_refused": 0,
+   "request_busy_s": 0.479,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pwahmzu/replay.mp4",
+  "duration_s": 0.04,
+  "fps": 25.0,
+  "frames": 1,
+  "sha256": "5e8d9ac5903805692bb091ec5982380b9ebf9aebb81126fec3fb06d75a0c882d"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第1次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第2次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第3次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第4次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第5次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第6次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第7次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第8次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第9次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第10次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第11次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第12次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第13次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第14次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第15次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第16次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第17次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第18次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第19次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频仅有视频 0.00 s 的初始帧（全片0.04秒、1帧）；机器人站立，红色球拍垂在右手旁，橙色球仍在释放器处，未出现拍面击球或随后上升，未达到第20次有效颠球。"
+     },
+     {
+      "t": null,
+      "log_step": 24,
+      "text_zh": "观察：log step 24 中机器人服务状态为 episode=1、control_steps=0、rows=0、running=false；该步仅调用 seconds=0 的观测。最终评分字段 n_actions=0、sim_steps=0、rows=0、clock_s=0，证实受评回合未推进，未执行任何颠球。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "不适用",
+      "how": "不适用",
+      "note_zh": "观察：评分记录 n_actions=0、sim_steps=0、rows=0；视频 0.00 s 仅为开局画面，没有交付执行轨迹，故无此步骤的实际操作接触可标注。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "unknown",
+   "t": null,
+   "log_step": 143,
+   "where": "首步前",
+   "evidence_zh": "观察：log step 143 最后一条消息明确记录“This request was blocked by Gemini's filters”，没有后续调用；此前 log step 142 仍在修改本地仿真测试脚本。log step 24 的机器人服务为 control_steps=0、running=false，最终评分字段 n_actions=0、sim_steps=0，与视频 0.00 s 唯一初始帧一致。推断：模型提供方过滤器在动作交付和物理步骤启动前中断会话，没有可定位的机器人失败步骤，依外部因素规则置 unknown／首步前。"
+  },
+  "phenomena": [],
+  "root_causes": [],
+  "gap_zh": "观察：inputs/replay.mp4 存在，但全片仅0.04秒、1帧，显示视频 0.00 s 初始状态；inputs/没有状态日志。log step 143 记录模型提供方 Gemini 过滤器阻断，评分记录确认从未提交动作或推进仿真。推断：这是结果尚未形成前的外部中断，应重试并排除于机器人失败分析；保留目录 success=0，不标“没尝试”，现象和根因均留空。本地仿真测试不属于受评回合，不能据其结果归因；评分字段亦未报告三个原生指标，指标值均留 null。",
+  "unmet": {
+   "bounce_1": false,
+   "bounce_2": false,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  }
+ },
+ "summary_zh": "观察：本次视频仅含视频 0.00 s 的初始帧，球仍在释放器处，最终评分记录动作数和仿真步数均为0，20个有效颠球节点均未达到。观察：log step 143 记录 Gemini 过滤器阻断，此前 agent 仍在本地仿真测试，未向受评服务提交动作。推断：本次属首步前的外部中断，断点为 unknown，应重试并排除于机器人失败分析；无执行轨迹和状态日志可支持物理失败现象或根因，保留目录 success=0。",
+ "conf": 10,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "标注与证据一致：视频仅1帧（视频0.00 s），评分字段 n_actions=0、sim_steps=0，log step 143 为 Gemini 过滤器中断，交付前被外部因素切断，现象和根因留空并有 gap_zh，20个节点均未达到，符合 FAIL-5/D37。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:02:48Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pwahmzu/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pwahmzu/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "b9e2498e44d0911ce6403fdb66c8ffd6ed1a328f437d94be1b380f419f57e8e1",
+   "video_sha256": "5e8d9ac5903805692bb091ec5982380b9ebf9aebb81126fec3fb06d75a0c882d",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pxpscgq.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pxpscgq.json
new file mode 100644
index 0000000000000000000000000000000000000000..30fde19e4324ce5e89083562e5f52a41f88057cb
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pxpscgq.json
@@ -0,0 +1,478 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pxpscgq",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 305,
+   "sim_steps": 1842,
+   "calls": 203,
+   "rows": 2038,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 594.0,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "realtime": 1,
+   "slowdown": 1.0,
+   "clock_s": 18.435,
+   "max_lag_s": 0.0137,
+   "stalls": 0,
+   "stalled_s": 0.0,
+   "stalled_share": 0.0,
+   "held_steps": 16,
+   "late_steps": 0,
+   "requests_refused": 0,
+   "request_busy_s": 2.251,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pxpscgq/replay.mp4",
+  "duration_s": 3.08,
+  "fps": 25.0,
+  "frames": 77,
+  "sha256": "fea21caf8f66c4bb7e46401f7f955f762e06ca84516a29d6a2f14424ae040b0a"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.48,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.32,
+      "log_step": null,
+      "text_zh": "观察：视频 2.28–2.36 s，橙色小球落到机器人右手所持红色拍面，右臂上击后小球向上离拍。"
+     },
+     {
+      "t": 2.48,
+      "log_step": 101,
+      "text_zh": "观察：视频 2.44–2.48 s，小球从拍面升回并超过开局释放杯中球心高度，随后在视频约 2.64–2.68 s 到达顶点；log step 101 中最后一局 episode 6 的服务裁判返回 stages_reached=1。该字段是机器人服务输出的有效颠球阶段数，确认第1次满足高度、速率及接触验收；达到高度的时间取视频约 2.48 s（误差约一帧），不是 agent 工作时间。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.28–2.36 s，机器人右手持球拍，以红色拍面击起橙色小球；候选方式“击打”，现有接触方式词表无击打项。手握拍柄属于同一工具链，不另计。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–3.04 s，机器人双脚持续在地面支撑身体，右臂持拍上击时未跌倒。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 2.76–3.04 s，小球从高处回落并向画面左侧、机器人右手方向偏移，最终到达右手／腕部上方，位于红色拍面的画面左侧，没有第二次拍面击起及达标上升。log step 101 的最后一局服务裁判返回 stages_reached=1、done=True，outcome 为“the ball touched something other than the paddle's blade”；第2次有效颠球未完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "其他",
+      "note_zh": "推断：视频 3.00–3.04 s，小球落到持拍右手／腕部上方而非红色拍面；结合 log step 101 服务裁判的非拍面接触判负，实际发生右手端附近的意外碰球。候选方式“意外碰撞”，不属于抓握、推扫或托舀；具体手部或腕部碰撞体未记录。右手虽持拍尝试续颠，本次未见工具拍面触球，故不填成功击打的工具接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.76–3.04 s，机器人双脚仍支撑地面，准备第二次持拍上击期间身体站立。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第3次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第3次未开始。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第4次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第4次未开始。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第5次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第5次未开始。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第6次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第6次未开始。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第7次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第7次未开始。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第8次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第8次未开始。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第9次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第9次未开始。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第10次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第10次未开始。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第11次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第11次未开始。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第12次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第12次未开始。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第13次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第13次未开始。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第14次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第14次未开始。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第15次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第15次未开始。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第16次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第16次未开始。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第17次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第17次未开始。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第18次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第18次未开始。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第19次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第19次未开始。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.04,
+      "log_step": 101,
+      "text_zh": "观察：视频 3.04 s 结束前只出现首球上击及其回落，未出现第20次有效颠球；log step 101 中最后一局 episode 6 的服务裁判已判负，stages_reached=1。第2次尚未完成，后续第20次未开始。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_2",
+   "t": 3.0,
+   "log_step": 101,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 2.76–3.04 s，机器人持拍等待并尝试承接回落球，但小球向红色拍面的画面左侧偏移，末帧落到右手／腕部附近，没有形成第二次拍面反弹。log step 100 重置为最后一局 episode 6；log step 101 执行该局后，服务返回 stages_reached=1，因球接触拍面以外物体而结束，之后未恢复。"
+  },
+  "unmet": {
+   "bounce_1": true,
+   "bounce_2": false,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 3.0,
+    "log_step": 101,
+    "evidence_zh": "观察：视频 2.76–3.04 s，小球回落过程中向持拍手一侧偏离，机器人红色拍面没有跟到回落球下方，小球在右手／腕部附近终止运动，未被拍面再次击起。log step 101 的服务裁判确认此次是非拍面触球并判负；机器人对运动中的球未完成第二次拦截。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "推断：log step 99 的最终控制脚本把击球轨迹预计算在固定水平位置，并用球像素偏差修正腕部倾角；log step 101 记录该脚本在最后一局运行并发出第二次上击指令。视频 2.48–3.04 s 显示首球反弹后横向偏移、回落到持拍手附近，拍面未成功追随拦截。由此将失误归于连续颠球的运动控制；现有数据不足以区分拍面倾角、轨迹跟踪和视觉反馈响应的贡献，不进一步指定子类。"
+   }
+  ],
+  "gap_zh": "无逐帧状态日志或接触对，视频只有场地视角且为25 fps，不能识别末次碰球的具体手／腕碰撞体，也不能独立测定毫秒接触时长及精确速度比；第1次有效性与后续未达由最后一局服务裁判的 stages_reached=1 补足。没有逐次阶段锁存时间，首节点达到时间按视频高度近似定位。run.json 的 reward 未报告 bounces、ball_height_m 或 grip_break_max_s 原生指标字段，native_metrics 均留 null。缺少实际拍面法向、球速度和完整反馈遥测，根因仅标控制，不细分倾角与响应延迟。"
+ },
+ "summary_zh": "观察：最后受评的是 log step 100 重置、log step 101 执行的 episode 6；视频中机器人右手持拍完成第1次有效颠球，视频约 2.48 s 球升过释放高度，服务裁判确认 stages_reached=1，最终进度为1/20。推断：第2次颠球在视频约 3.00 s 失误，回落球偏向持拍手，拍面未跟上，视频 3.04 s 球位于右手／腕部附近且服务判为非拍面接触，属于动态拦截的控制失误。缺少接触对与拍面、球运动遥测，无法确定具体碰撞体或进一步细分控制原因；原生指标字段未报告。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "视频3.08 s即受评的episode 6（305个控制步≈3.05 s），首次有效颠球约2.48 s、第2次在约3.0 s因球落到手/腕而非拍面失败，与log step 101的stages_reached=1一致；节点、断点、现象和根因均有证据支持，未发现需要修改的内容。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:04:06Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pxpscgq/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--pxpscgq/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "9da491a9f0c1c47b2fa54ad9c3bf986445ddeeaac3466308610d2affb556d36a",
+   "video_sha256": "fea21caf8f66c4bb7e46401f7f955f762e06ca84516a29d6a2f14424ae040b0a",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--qjeahnv.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--qjeahnv.json
new file mode 100644
index 0000000000000000000000000000000000000000..23bbcf902c03d582dfb23611c2cb5625e38d8b24
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--qjeahnv.json
@@ -0,0 +1,485 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard-realtime--qjeahnv",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 252,
+   "sim_steps": 1419,
+   "calls": 26,
+   "rows": 1350,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 495.4,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "realtime": 1,
+   "slowdown": 1.0,
+   "clock_s": 14.182,
+   "max_lag_s": 0.0184,
+   "stalls": 0,
+   "stalled_s": 0.0,
+   "stalled_share": 0.0,
+   "held_steps": 176,
+   "late_steps": 0,
+   "requests_refused": 1,
+   "request_busy_s": 1.387,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--qjeahnv/replay.mp4",
+  "duration_s": 2.56,
+  "fps": 25.0,
+  "frames": 64,
+  "sha256": "bd0c1e1c4c1b25f07305ce7d68fd117879353212f69ea6660e909ed681384231"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.44,
+      "log_step": 49,
+      "text_zh": "观察：视频0.00–1.44 s，机器人右手持红色球拍转动至较平的姿态，双脚保持站立；视频1.44–2.00 s拍面留在身体旁，球仍在释放杯中。log step 49的最后一次reset及submit代码提交了右臂预设关节轨迹，视频中的调姿与该尝试相符。"
+     },
+     {
+      "t": 2.32,
+      "log_step": null,
+      "text_zh": "观察：视频2.04–2.32 s，橙色球离开释放杯后持续下落，从红色拍面的画面右侧经过，未出现触拍后向上反弹；视频2.36–2.52 s球继续降至地面附近。全片未见一次有效颠球。"
+     },
+     {
+      "t": 2.52,
+      "log_step": 52,
+      "text_zh": "观察：log step 52中agent自述最后一次尝试在control step 252以“the ball touched the floor”结束、0/20次颠球；该说法与视频末段相符。节点判断依据为视频中的漏接，未将会话前几次重置的结果计入。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.00–2.52 s，机器人双脚在地面支撑身体，同时右臂持拍调姿。视频2.04–2.52 s未见拍面触球，因此不将任务典型方案中的“击打”记为已发生接触；持工具的手与拍柄不重复记作工具—球接触。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第2次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第3次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第4次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第5次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第6次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第7次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第8次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第9次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第10次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第11次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第12次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第13次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第14次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第15次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第16次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第17次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第18次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第19次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.52,
+      "log_step": null,
+      "text_zh": "推断：视频0.00–2.52 s中球先留在释放杯内，再于视频约2.04 s起下落，漏过球拍后抵达地面附近，全程没有触拍上升；第1次有效颠球未达成，按DAG逐次计数依赖，第20次有效颠球亦未达成且未开始单独尝试。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_1",
+   "t": 2.32,
+   "log_step": 49,
+   "where": "操作中",
+   "evidence_zh": "观察：视频0.00–1.44 s机器人已右手持拍调姿，开始为首次颠球预置；视频2.24–2.32 s下落小球从拍面旁漏过，视频2.48 s球到地面附近，此后未修复。log step 49提交的是最后一次重置后的持拍轨迹，log step 51保存同一组关节目标；断点为正在尝试的第1次颠球。"
+  },
+  "unmet": {
+   "bounce_1": false,
+   "bounce_2": false,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 2.32,
+    "log_step": null,
+    "evidence_zh": "观察：视频1.44–2.52 s，机器人右手持红色拍面基本保持在身体旁；视频2.04–2.32 s橙色球离开释放杯后持续下落，从拍面的画面右侧经过，未见触拍后向上反弹；视频2.36–2.52 s球继续降至地面附近。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 2.32,
+    "log_step": 49,
+    "evidence_zh": "推断：log step 49的控制脚本将右臂插值到固定目标关节姿态后保持；视频1.44–2.52 s拍面基本停在身体旁，视频2.04–2.32 s落球从拍面侧旁经过，说明该次持拍定位控制没有使拍面与落球路径重合。log step 44–45记录了agent自行建模的FK与目标位姿估算，但没有最后一局的实际关节跟踪及球拍三维位姿日志，不能区分目标位姿计算或几何标定偏差与执行跟踪误差，根因保留为控制，子类未定。",
+    "source_zh": "日志"
+   }
+  ],
+  "gap_zh": "未提供状态日志；最后一局的服务结果在log step 49输出中未保留，0/20和触地终止来自agent在log step 50–52的自述，已用视频漏接、落地过程交叉核对。缺少最后一局实际关节跟踪和球拍三维位姿，无法区分目标位姿计算误差与执行跟踪误差，根因保留为控制。run.json的reward未报告bounces、ball_height_m、grip_break_max_s，三个原生指标均置null；不将agent自述的计数冒充评分器指标。"
+ },
+ "summary_zh": "受评录像为最后一次重置后的持拍尝试：机器人调平球拍后等待落球，但视频约2.32 s小球从拍面旁漏过，首次颠球失败，20个节点均未达成。固定目标姿态未使拍面与落球路径重合，支持持拍定位控制不足；没有状态日志及最后一局关节跟踪记录，不能细分位姿计算、几何标定与执行误差。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 7,
+    "summary_zh": "节点（20个均未达成）、断点bounce_1、时钟均无问题。现象标签有误：球从释放杯自由落下，落点固定，拍面全程静止停在身体旁，球从拍面侧旁漏过，是拍面位姿没放到球的落点下方（位姿不准），不是对运动目标的时机/跟随问题。",
+    "n_issues": 2,
+    "n_major": 1
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "20个节点均未达成、断点bounce_1、时钟与证据均与视频一致；球自释放杯下落并从静止拍面旁漏过，未见触拍反弹。动态时机依CASES FAIL-14和03文档对运动目标的规定可保留，根因停在控制并如实说明不能细分。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:06:22Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--qjeahnv/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--qjeahnv/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--qjeahnv/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--qjeahnv/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "cd28e55a784fa0b0cd987ecf9dd7a37dc9707d578b752962d72b8ec60eda2447",
+   "video_sha256": "bd0c1e1c4c1b25f07305ce7d68fd117879353212f69ea6660e909ed681384231",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--xnpomdk.json b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--xnpomdk.json
new file mode 100644
index 0000000000000000000000000000000000000000..d350c50af2f13981a634350bd41edbd4f327b180
--- /dev/null
+++ b/phase2/embodied-olympics-fun/paddle_bounce/trajectories/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--xnpomdk.json
@@ -0,0 +1,492 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-paddle-bounce-i00-standard-realtime--xnpomdk",
+ "taskKey": "embodied-olympics-fun/paddle_bounce",
+ "dag_sha256": "0430fbd3fb280dd4add17cdc2f794688f20fdd9ca4d316d2240c5df644bbea14",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 369,
+   "sim_steps": 1723,
+   "calls": 1061,
+   "rows": 8585,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 1863.1,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "realtime": 1,
+   "slowdown": 1.0,
+   "clock_s": 18.197,
+   "max_lag_s": 0.4176,
+   "stalls": 0,
+   "stalled_s": 0.0,
+   "stalled_share": 0.0,
+   "held_steps": 72,
+   "late_steps": 342,
+   "requests_refused": 4,
+   "request_busy_s": 3.746,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--xnpomdk/replay.mp4",
+  "duration_s": 3.72,
+  "fps": 25.0,
+  "frames": 93,
+  "sha256": "3d8a2f59619589cf66bbe8c780fa327a208cec9d0bdd41ab6beb50e6ac3ebe5b"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 20,
+  "base": 0,
+  "peak": 2,
+  "final": 2,
+  "nodes": {
+   "bounce_1": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.4,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.4,
+      "log_step": 110,
+      "text_zh": "观察：视频 2.20–2.28 s，机器人右手持红色球拍触及下落的橙球，随后视频 2.36–2.52 s球明显离拍上升，回到原释放杯高度附近或更高；全片依次可见两次这样的高反弹。log step 110 在最后一次 reset 后查询的是实时服务的裁判结果，stages_reached=2、stages_total=20，确认前两次是有效颠球。完成时刻取上述上升帧，约有0.1秒不确定性；毫秒接触与速度比由服务的有效阶段计数确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.20–2.28 s，机器人右手持红色球拍，拍面接触橙球，随后球离拍上升。候选接触方式为“击打”，现有方式词表无对应项，故用其他；不重复标手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.20–3.68 s，机器人双脚持续落在地面上，支撑持拍击球动作，未见跌倒。"
+     }
+    ]
+   },
+   "bounce_2": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.92,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 2.92,
+      "log_step": 110,
+      "text_zh": "观察：视频 2.72–2.80 s，机器人右手持红色球拍触及下落的橙球，随后视频 2.88–3.04 s球明显离拍上升，回到原释放杯高度附近或更高；全片依次可见两次这样的高反弹。log step 110 在最后一次 reset 后查询的是实时服务的裁判结果，stages_reached=2、stages_total=20，确认前两次是有效颠球。完成时刻取上述上升帧，约有0.1秒不确定性；毫秒接触与速度比由服务的有效阶段计数确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 2.72–2.80 s，机器人右手持红色球拍，拍面接触橙球，随后球离拍上升。候选接触方式为“击打”，现有方式词表无对应项，故用其他；不重复标手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.20–3.68 s，机器人双脚持续落在地面上，支撑持拍击球动作，未见跌倒。"
+     }
+    ]
+   },
+   "bounce_3": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.48,
+      "log_step": 110,
+      "text_zh": "观察：视频 3.24–3.28 s，橙球接近机器人右手所持红拍后再次上升；视频 3.40–3.52 s 的最高位置低于前两轮的最高位置，视频 3.56–3.68 s 球回落到拍面，未出现第三次回到释放高度的合格反弹。log step 110 的最后一局服务结果为2/20，终止原因为“the ball came back to the blade without having risen enough”，确认第3次有效颠球未完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手持工具",
+      "how": "其他",
+      "note_zh": "观察：视频 3.24–3.28 s，机器人右手持红色球拍，拍面接触橙球，随后球离拍上升。候选接触方式为“击打”，现有方式词表无对应项，故用其他；不重复标手握拍柄。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.20–3.68 s，机器人双脚持续落在地面上，支撑持拍击球动作，未见跌倒。"
+     }
+    ]
+   },
+   "bounce_4": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第4次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_5": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第5次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_6": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第6次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_7": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第7次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_8": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第8次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_9": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第9次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_10": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第10次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_11": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第11次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_12": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第12次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_13": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第13次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_14": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第14次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_15": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第15次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_16": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第16次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_17": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第17次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_18": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第18次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_19": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第19次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   },
+   "bounce_20": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.68,
+      "log_step": 110,
+      "text_zh": "观察：视频 0–3.68 s 中机器人仅完成前两次高反弹，第三轮低反弹后画面结束；log step 110 确认受评末局以 stages_reached=2/20 和低弹回拍的原因终止。推断：第20次有效颠球的顺序前置尚未完成，该节点未达到，也未单独开始尝试。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "bounces",
+   "unit": "count",
+   "threshold": 20,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "ball_height_m",
+   "unit": "m",
+   "threshold": 1.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "grip_break_max_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "bounce_3",
+   "t": 3.28,
+   "log_step": 110,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 3.24–3.28 s，机器人右手持拍开始第三轮反弹；视频 3.40–3.52 s 球的顶点偏低，随后视频 3.68 s 回到拍面。log step 110 明确先执行最后一次 reset，再运行控制脚本并查询裁判，服务报告已完成2/20，因球升高不足就回拍而终止。推断：首次未修复的失败是第3次有效颠球，导航锚点取这次击球后刚开始上升处。"
+  },
+  "unmet": {
+   "bounce_1": true,
+   "bounce_2": true,
+   "bounce_3": false,
+   "bounce_4": false,
+   "bounce_5": false,
+   "bounce_6": false,
+   "bounce_7": false,
+   "bounce_8": false,
+   "bounce_9": false,
+   "bounce_10": false,
+   "bounce_11": false,
+   "bounce_12": false,
+   "bounce_13": false,
+   "bounce_14": false,
+   "bounce_15": false,
+   "bounce_16": false,
+   "bounce_17": false,
+   "bounce_18": false,
+   "bounce_19": false,
+   "bounce_20": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 3.48,
+    "log_step": 110,
+    "evidence_zh": "观察：视频 3.24–3.52 s，机器人右手球拍确实将球弹起，但这轮球的顶点低于前两轮、未回到原释放高度，随后视频 3.56–3.68 s 回落到拍面；log step 110 的实时裁判以升高不足后回拍判负。推断：击出的球未达到要求的竖直目标位置，表现为颠球高度偏低；没有把已建立的触拍误标为没建立接触。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：log step 109 中 agent 的脚本把目标顶点 ZA 设为1.28米，目标高于1.20米验收高度；但视频 3.24–3.68 s 的实际第三轮反弹偏低，并由 log step 110 的服务裁判确认升高不足。log step 111 打印该末局脚本的视觉拟合，第三轮预测顶点约1.178米，而前两轮约1.279、1.280米（这些是脚本估计，并非裁判真值）。结合前两轮成功建立拍球接触，支持击球控制未产生需要的上升结果；缺少实际拍面速度、姿态误差与接触冲量，不能进一步指定控制子类或认定某参数错误。",
+    "source_zh": "视频 / 日志",
+    "t": 3.28,
+    "log_step": 111
+   }
+  ],
+  "gap_zh": "没有状态日志或逐次裁判锁存时刻，节点完成时刻由视频上升帧近似定位；25 fps画面不能独立复核0.02秒接触、0.98速度比或0.05秒握拍中断，由末局服务的2/20阶段结果确认有效次数。没有末局实际拍面速度、姿态跟踪误差和接触力数据，无法区分击球速度不足、拍面倾斜或跟踪误差；不沿用前五局的感知或脚本问题作为本局原因。reward未报告DAG所列三个原生指标字段，因此native_metrics均保留null，2/20来自log step 110的服务阶段字段。"
+ },
+ "summary_zh": "受评的是 log step 110 最后一次重置后的第6局；视频约2.40 s、2.92 s 完成前两次有效颠球，末态为2/20，其余节点未完成。视频约3.28 s 开始的第三轮反弹偏低，随后回拍，服务裁判以升高不足判负；归因为击球控制未产生所需上升结果。缺少逐次裁判时刻及拍面速度、姿态和接触力数据，无法细分控制原因。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "前两次有效颠球（视频约2.4 s、2.92 s）与日志 run_log 事件（hit 2.22/2.738 s，adapt 顶点1.279/1.28 m）及服务2/20一致；第3次顶点1.178 m偏低、断点bounce_3、控制类归因均有证据支撑，时钟与节点值未见错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:02:57Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--xnpomdk/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-paddle-bounce-i00-standard-realtime--xnpomdk/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "756d255a22fde5fb5aeccec27003a0e593f3799665ea5a84200aa5efbd818f2c",
+   "video_sha256": "3d8a2f59619589cf66bbe8c780fa327a208cec9d0bdd41ab6beb50e6ac3ebe5b",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--5x3ovwh.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--5x3ovwh.json
new file mode 100644
index 0000000000000000000000000000000000000000..ef00a2a8ae120697dba7b006015024b0ee7b0dbe
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--5x3ovwh.json
@@ -0,0 +1,226 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-privileged--5x3ovwh",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.3333,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 1881,
+   "sim_steps": 1881,
+   "horizon": 6000,
+   "episode_over": 0,
+   "progress": 0.3333,
+   "stages_reached": 1,
+   "stages_total": 3,
+   "fell": 0,
+   "replay_s": 9.55,
+   "bad_trajectory": 0,
+   "metric_rear_x_m": 10.426,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 2248,
+   "agent_resets": 3,
+   "agent_replays": 5,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 2.918,
+   "peak_processes": 30,
+   "peak_disk_gb": 0.988,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-privileged--5x3ovwh/replay.mp4",
+  "duration_s": 18.84,
+  "fps": 25.0,
+  "frames": 471,
+  "sha256": "a508af090a949703f38b9ce96e08c38b839f1ee96f013933056ee51466156ce8"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [
+     [
+      17.0,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 17.0,
+      "log_step": null,
+      "text_zh": "观察：视频 14.0–16.0 s，机器人步行接近白色横杆，屈腿、起跳并再次蹬地；视频 16.5–17.1 s，机器人身体翻转，两脚向上，骨盆及两脚先后越到横杆后的蓝色坑一侧，约视频 17.0 s 全身清栏。时间由逐帧画面估计，误差约 0.2 s。"
+     },
+     {
+      "t": 17.0,
+      "log_step": 256,
+      "text_zh": "观察：log step 256 对最终交付的 1881 行轨迹重新生成并回放，文件哈希保持一致，原生裁判返回 stages=[\"hurdle\"]、progress=0.3333；log step 240 的两次回放终态哈希相同，评分器字段 deterministic=1、stages_reached=1。原生 hurdle 阶段确认精确的三点过栏及全身清栏验收；本节点按事件保留。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.0–15.9 s，机器人双脚交替接触红色跑道行进，随后屈腿蹬地、落回栏前跑道并再次蹬地越栏；作用端为脚部，对象为栏前跑道。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 17.2,
+      "log_step": null,
+      "text_zh": "观察：视频 17.0–17.5 s，机器人两脚仍朝上、上身向坑底落下，随后身体低伏在蓝色坑近端；视频 17.8–18.8 s 的逐帧画面中没有重新站起，也未沿坡面到达坑远端红色跑道。"
+     },
+     {
+      "t": 18.8,
+      "log_step": 256,
+      "text_zh": "观察：log step 256 最终交付回放的原生裁判仅报告 stages=[\"hurdle\"]、rear_x_m=10.426、success=false、failed=null、episode_over=false，评分器同样报告 metric_rear_x_m=10.426，尚未达到全身出坑要求的 13.66 m。"
+     },
+     {
+      "t": 18.8,
+      "log_step": null,
+      "text_zh": "推断：本次录像内机器人从未完成出坑并站稳；成功参考录像在其视频 22.0–26.5 s 可见机器人起身、穿过坑并站到远端跑道，同一全景中的这一可见结果在本次视频始终没有出现。参考录像只用于辨认验收外观，不作为本次进度证据。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "推断：视频 17.2–17.7 s，机器人倒置落入干坑后身体低伏并停止下落，说明躯干等非脚部部位由坑底承托；视频 18.8 s 仍未建立直立的脚部支撑。坑沿遮住部分着地点，不能逐一确认手臂与腿的接触对。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "value": 10.426,
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better"
+  },
+  {
+   "name": "time_s",
+   "value": null,
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "pit_crossed_standing",
+   "t": 17.2,
+   "log_step": 179,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 17.0 s 机器人已清栏，视频 17.1–17.5 s 随即倒置落入坑内，未形成脚朝下的落地支撑，直到视频 18.8 s 仍低伏在坑近端。log step 179 的交付脚本在越栏后执行伸腿、缓冲及起身关节目标，故此时正在尝试出坑站稳步骤；越栏事件已经完成，后续出坑站稳始终没有补成。"
+  },
+  "unmet": {
+   "hurdle_cleared": true,
+   "pit_crossed_standing": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 17.2,
+    "log_step": null,
+    "evidence_zh": "观察：视频 16.8–17.1 s 机器人双脚高于骨盆、上身倒置；视频 17.2–17.7 s 机器人上身落入坑内并低伏，视频 17.8–18.8 s 未恢复直立脚部支撑。这是翻倒后的落地失衡；评分器汇总 fell=0 不改变画面显示的翻倒事实。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 17.2,
+    "log_step": 179,
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "推断：log step 179 的最终生成脚本以屈腿、伸腿、收腿、再次伸腿及落地缓冲的关节目标执行越障；log step 256 重生成文件哈希不变并复现仅 hurdle 阶段。结合视频 16.5–17.7 s 的持续翻转和倒置落地，这套关节动作未能控制越栏后的全身姿态并建立脚部支撑，导致无法继续出坑。agent 在 log step 263 也自述未解决旋转及落坑后的起身；该自述仅作辅助，不把其他会话试验视为本次轨迹。"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志；固定全景中机器人较小，坑沿遮住部分落地接触，无法核实每个着地部位、具体关节误差和失衡的控制子类，根因仅标到控制。越栏的精确高度与全身边界由最终交付回放的原生 hurdle 阶段支持，视频达到时刻为约值。原生 time_s 完成成绩未在评分器 reward 中报告，指标保留 null；log step 256 的 time_s=18.81 是未完成轨迹的回放长度。"
+ },
+ "summary_zh": "机器人完成越栏，两个语义节点最终达到 1/2；原生裁判的 1/3 表示三个内部阶段中仅 hurdle 完成。机器人约在视频 17.0 s 清栏，随后倒置落入坑内，直到视频 18.8 s 仍未起身或出坑，断在出坑并站稳，现象为平衡，证据支持根因定位到控制。最终交付多次回放一致；无逐帧状态日志且落地接触部分被遮挡，无法进一步确定具体控制子类。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标注与视频、日志一致：约视频 17.0 s 清栏（评分器 stages=[hurdle]），随后倒置落入坑内，至 18.8 s 未出坑；断点、平衡现象与控制根因有证据支持，未发现需改动之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:52:44Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--5x3ovwh/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--5x3ovwh/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "45f0ec59dd3d9c7b18093ecf132832cf0b53b26b8a14cd69b6d1c163b6aa0d12",
+   "video_sha256": "a508af090a949703f38b9ce96e08c38b839f1ee96f013933056ee51466156ce8",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--ewbe3mt.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--ewbe3mt.json
new file mode 100644
index 0000000000000000000000000000000000000000..be4dd179f83430ed78cf83c7ddbd7a3c98d36125
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--ewbe3mt.json
@@ -0,0 +1,175 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-privileged--ewbe3mt",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 10,
+   "sim_steps": 10,
+   "horizon": 6000,
+   "episode_over": 0,
+   "progress": 0.0,
+   "stages_reached": 0,
+   "stages_total": 3,
+   "fell": 0,
+   "replay_s": 2.26,
+   "bad_trajectory": 0,
+   "metric_rear_x_m": -0.11,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 17,
+   "agent_resets": 178,
+   "agent_replays": 0,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.763,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.256,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-privileged--ewbe3mt/replay.mp4",
+  "duration_s": 0.12,
+  "fps": 25.0,
+  "frames": 3,
+  "sha256": "b16b4d2975900b01f32ff6e4daf3f978273acae4cfceb2a3fabd1773c3c8cce9"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.08,
+      "log_step": 10,
+      "text_zh": "观察：已逐帧查看本次视频 0.00、0.04、0.08 s，机器人始终直立于左侧起点附近，白色障碍栏远在其右前方，没有接近或越栏。log step 10 的脚本仅执行十步 s.hold() 并保存，备注明确为初始占位轨迹；评分器 n_actions=10、stages_reached=0、metric_rear_x_m=-0.11，与画面一致。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.08,
+      "log_step": null,
+      "text_zh": "观察：本次视频 0.00、0.04、0.08 s 中，机器人一直在白栏前方起点，未进入蓝色水坑，更未到达远岸红色跑道。成功参考视频约 25.8–26.8 s 可见机器人全身在坑远端站立，本次全部画面均没有该位置变化；评分器 stages_reached=0、metric_rear_x_m=-0.11，亦未达到出坑位置。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better",
+   "value": -0.11
+  },
+  {
+   "name": "time_s",
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "unknown",
+   "t": null,
+   "log_step": 138,
+   "where": "首步前",
+   "evidence_zh": "观察：本次视频 0.00–0.08 s 仅见机器人留在起点，未启动越栏；log step 10 保存的是十步原地保持的占位文件。log step 137 仍在独立世界中测试，且没有保存新轨迹；log step 138 的末条回复为模型提供方过滤阻断提示。评分器 agent_replays=0。推断：正式解尚未保存并验证即被外因截断，没有可定位的评分局物理失败步骤；依外因规则保留 unknown / 首步前。"
+  },
+  "unmet": {
+   "hurdle_cleared": false,
+   "pit_crossed_standing": false
+  },
+  "phenomena": [],
+  "root_causes": [],
+  "gap_zh": "观察：inputs/ 有完整三帧评分视频和运行日志，没有逐帧状态日志。log step 10 仅保存初始占位文件；log step 137 通过直接重放置机器人到栏前开展独立测试并打印 succ=True，但未保存或验证新交付；log step 138 以模型提供方过滤提示结束，评分器 agent_replays=0。推断：这不是已保存并验证正式交付后的过滤，按外因截断规则（CASES.md FAIL-5、DECISIONS.md D37）不归为机器人失败，故不标没尝试、控制或交付根因；会话测试中的接触、碰栏和摔倒也不能移作本次评分局现象。两个节点在占位评分视频中均未开展，省略 contact_actual。"
+ },
+ "summary_zh": "评分视频全部三帧（视频 0.00、0.04、0.08 s）仅显示机器人在起点附近保持站立，越栏及出坑并站稳均未完成，终态为 0/2。log step 10 保存的十步占位轨迹与评分回放一致；log step 137 的独立测试成功不属于该交付。log step 138 在正式解保存并验证之前被模型提供方过滤截断，按外因规则将断点记为 unknown / 首步前，现象与根因留空。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "标注与证据一致：评分视频仅三帧，机器人原地站立，两节点均未达成；log step 10 为十步 hold 占位轨迹，step 137 为独立测试，step 138 为提供方过滤截断，按外因规则断点 unknown/首步前、现象根因留空并写 gap_zh，合理。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:49:30Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--ewbe3mt/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--ewbe3mt/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "99830abef6f3c154a5bb29aa731de15de18a91d6843d52a015cb00969ded1441",
+   "video_sha256": "b16b4d2975900b01f32ff6e4daf3f978273acae4cfceb2a3fabd1773c3c8cce9",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--f9wemhd.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--f9wemhd.json
new file mode 100644
index 0000000000000000000000000000000000000000..d08459494383ecea97ae44bc8056a6e3cb48e4a2
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--f9wemhd.json
@@ -0,0 +1,213 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-privileged--f9wemhd",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 3168,
+   "sim_steps": 3168,
+   "horizon": 6000,
+   "episode_over": 1,
+   "progress": 0.0,
+   "stages_reached": 0,
+   "stages_total": 3,
+   "fell": 0,
+   "replay_s": 14.04,
+   "bad_trajectory": 0,
+   "metric_rear_x_m": 9.612,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 61,
+   "agent_resets": 0,
+   "agent_replays": 2,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.777,
+   "peak_processes": 14,
+   "peak_disk_gb": 0.384,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-privileged--f9wemhd/replay.mp4",
+  "duration_s": 31.72,
+  "fps": 25.0,
+  "frames": 793,
+  "sha256": "bca0b0eb6dd02d30abb2053df8f76b24dccf4a96ae25ab2911bc1b071cde54e0"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 14.0,
+      "log_step": null,
+      "text_zh": "观察：视频0.0–14.0 s机器人沿红色跑道走向白色横杆，视频16.0–29.8 s仍在栏前调整和站立。"
+     },
+     {
+      "t": 31.2,
+      "log_step": null,
+      "text_zh": "观察：视频30.0–30.4 s机器人下蹲，视频30.6–31.0 s蹬地起跳并前倾，视频31.2–31.68 s下降，脚部仍低于白色横杆，未出现全身越到蓝色坑一侧的画面。"
+     },
+     {
+      "t": 31.68,
+      "log_step": 107,
+      "text_zh": "观察：log step 106生成并保存最终trajectory.npz；log step 107对该文件的新进程回放执行3168行，返回success=False、failed=\"did not go over the hurdle\"、rear_x_m=9.612；run.json评分器字段stages_reached=0、deterministic=1，与视频末帧仍未清栏一致。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：机器人双脚在栏前红色跑道上交替支撑行走，随后以脚部支撑下蹲并蹬地起跳；视频31.4–31.68 s脚部下降回栏前低处。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 31.68,
+      "log_step": 107,
+      "text_zh": "观察：视频0.0–31.68 s机器人始终未全身越栏，也未沿蓝色坑移动到远岸红色跑道；末帧仍在横杆近侧，出坑并站稳的可见结果从未出现。log step 107最终文件回放rear_x_m=9.612，run.json评分器stages_reached=0。"
+     },
+     {
+      "t": 31.68,
+      "log_step": null,
+      "text_zh": "推断：越栏前置事件未完成，且机器人未到达水坑远岸，因此“出坑并站稳”未开始、未达成；成功参考录像在其视频18.48 s已进入坑侧、其视频26.0–26.92 s已全身到达远岸，两种可见结果在本次录像中均不存在。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "value": 9.612,
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better"
+  },
+  {
+   "name": "time_s",
+   "value": null,
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "hurdle_cleared",
+   "t": 31.2,
+   "log_step": 107,
+   "where": "操作中",
+   "evidence_zh": "观察：视频30.6–31.0 s机器人从栏前起跳，但视频31.2–31.68 s脚部已下降至横杆以下，身体仍未全身清栏；log step 107最终交付文件回放被裁判以“did not go over the hurdle”判负，rear_x_m=9.612。此前没有完成后又丢失的节点。"
+  },
+  "unmet": {
+   "hurdle_cleared": false,
+   "pit_crossed_standing": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "步态",
+    "t": 31.2,
+    "log_step": 107,
+    "evidence_zh": "观察：视频30.6–31.0 s机器人确实起跳，但视频31.2–31.68 s在尚未全身越过白色横杆时就下降，脚部回到低处；录像结束前未见全身摔倒，run.json评分器fell=0。log step 107最终轨迹回放确认越栏失败。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "log_step": 106,
+    "evidence_zh": "推断：agent在log step 105将控制器改为折体抬脚、上举手臂并施加髋膝前馈力矩，log step 106用该版本生成最终交付轨迹；该生成器输出起跳vx=1.06 m/s、vz=2.34 m/s及末段骨盆x=9.87 m。结合视频30.6–31.68 s的起跳、下降和未清栏，以及log step 107的最终回放失败，所实现的蹬地与空中腿部运动不足以完成要求的越栏轨迹，归为控制；证据不足以区分轨迹设计、跟踪反馈或具体参数问题。",
+    "source_zh": "日志 / 视频 / 评分器字段"
+   }
+  ],
+  "gap_zh": "无逐帧状态日志，固定全景中机器人较小，且末段部分肢体与横杆重叠，不能确定骨盆或哪只脚在何高度触发过面判负，也不能确认是否存在短暂手撑栏；实际接触只记录可见的足部支撑。日志中agent自述的跳高机械极限未被独立验证，不能据此归因为任务或环境，也不能确定控制子类。"
+ },
+ "summary_zh": "观察：最终交付录像中，机器人走到栏前并下蹲起跳，视频31.2–31.68 s在未全身清栏时下降；越过障碍栏和出坑并站稳均未完成，最终为0/2。推断：断点为越栏操作中的步态不足，生成器实现的起跳与空中腿部控制未能达到清栏要求；无逐帧状态日志，具体触发判负的身体点、短暂栏上接触及控制子类无法确定。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "两个节点均未达成（0/2）与视频、log step 106/107（rear_x_m=9.612、did not go over the hurdle）及评分器字段一致；断点hurdle_cleared、步态现象与控制根因（已声明无法区分子类）有证据支持，未发现需要修改的内容。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:50:21Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--f9wemhd/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--f9wemhd/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "2f77b3339eadf3c91e25fe204466fd4d06511cb9a1b0d179a7ec4f47a2a281c5",
+   "video_sha256": "bca0b0eb6dd02d30abb2053df8f76b24dccf4a96ae25ab2911bc1b071cde54e0",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--kryaaa7.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--kryaaa7.json
new file mode 100644
index 0000000000000000000000000000000000000000..cd754d0c87b7b0d54c1236af089f476bbe1cf757
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--kryaaa7.json
@@ -0,0 +1,209 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-privileged--kryaaa7",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 1520,
+   "sim_steps": 1520,
+   "horizon": 6000,
+   "episode_over": 0,
+   "progress": 0.0,
+   "stages_reached": 0,
+   "stages_total": 3,
+   "fell": 0,
+   "replay_s": 11.34,
+   "bad_trajectory": 0,
+   "metric_rear_x_m": 9.447,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 416,
+   "agent_resets": 57,
+   "agent_replays": 1,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.117,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.3,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 1,
+   "session_log_from_monitor": 1
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-privileged--kryaaa7/replay.mp4",
+  "duration_s": 15.24,
+  "fps": 25.0,
+  "frames": 381,
+  "sha256": "16cc0dc34ce70930e9fbe311a16f73bd9dba63c015040ff91c11446ee802f8ee"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频从机器人站在红色跑道、白色障碍栏前方开始；视频全程的逐秒帧显示机器人向栏杆步行接近。"
+     },
+     {
+      "t": 15.2,
+      "log_step": 500,
+      "text_zh": "观察：视频14.0–15.2 s的加密帧显示机器人仍直立在白色横杆近侧，两脚仍在栏前跑道，未见骨盆与两脚翻越横杆；视频15.2 s最后一帧仍未完成越栏。log step 500的脚本仅回放并保存1520行接近栏杆的动作，log step 506所写生成脚本也只执行1520步walk；评分器报告stages_reached=0、metric_rear_x_m=9.447，支持这一可见结果。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.0–15.2 s，机器人左右脚交替接触栏前红色跑道，支撑身体向障碍栏步行接近。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 15.2,
+      "log_step": null,
+      "text_zh": "观察：本次视频0.0–15.2 s，机器人始终留在白色栏杆近侧，未进入蓝色水坑，更未到达远岸红色跑道。成功参考视频18.0–19.0 s可见机器人越到坑内，参考视频26.0–26.92 s可见其在远岸直立；本次全景没有出现这些位置变化，故此后续节点未达到也未开始。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "value": 9.447,
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better"
+  },
+  {
+   "name": "time_s",
+   "value": null,
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "hurdle_cleared",
+   "t": 15.0,
+   "log_step": 500,
+   "where": "操作中",
+   "evidence_zh": "观察：视频14.0–15.2 s，机器人步行到白色栏杆前，身体保持直立，但两脚与骨盆未越到坑侧，交付轨迹即结束。log step 500保存的是1520步接近轨迹，log step 506的生成脚本只有walk循环；评分器n_actions=1520、episode_over=0、fell=0、stages_reached=0，说明记录结束时越栏尚未完成。视频15.0 s仅作为未完成越栏的片尾导航锚点，未观察到单独的摔倒或违规终止事件。"
+  },
+  "unmet": {
+   "hurdle_cleared": false,
+   "pit_crossed_standing": false
+  },
+  "phenomena": [],
+  "root_causes": [],
+  "gap_zh": "无逐帧状态日志。交付轨迹止于向障碍栏接近的1520步walk（log step 500、506；评分器n_actions=1520），视频0.0–15.2 s及视频14.0–15.2 s逐帧检查均未见跳跨或撑栏越障动作，也未见摔倒，故不能将未执行的越栏动作标为步态失效；依CASES.md FAIL-23，现象层留空。保留越栏步骤中接近阶段的断点及双脚对跑道的支撑接触，不将接近行走等同于机器人从未开始该步工作。log step 52、107的试验分别在仿真1530、1645个控制步失败，不属于交付的1520步评分轨迹；docs/03_trajectory.md Step 3允许会话试验最多作为根因证据，但这两次失败不足以证明交付为何止于接近段，也不能由未交付越栏动作反推控制根因，故根因层留空。片尾固定全景中拳垫与横杆投影接近、重叠，无法可靠确认是否实际触栏或承重，故contact_actual仅记录可见的双脚对跑道支撑。"
+ },
+ "summary_zh": "评分轨迹仅使机器人从起点步行接近白色障碍栏，视频15.2 s仍留在栏前，越栏与出坑并站稳均未完成（0/2）。断点保留在越过障碍栏步骤的接近阶段；交付不含越栏动作，无法据此认定步态失效，会话试验也不足以确定交付止于接近段的根因，因此现象和根因均留空。无逐帧状态日志，片尾拳垫是否触栏不能确定。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 7,
+    "summary_zh": "节点值（0/2）、时间和接触与视频一致；问题在断点之后：交付轨迹只含1520步接近行走、没有任何越栏动作，却标了“步态”现象和“控制”根因，后者靠的是会话内试验（非评分回合）。",
+    "n_issues": 2,
+    "n_major": 2
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点0/2、断点hurdle_cleared、接触（双脚支撑）与视频一致；交付轨迹只含1520步接近行走（log 500/506），现象与根因留空并写gap_zh符合FAIL-23，第1轮意见已正确采纳。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:01:49Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--kryaaa7/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--kryaaa7/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--kryaaa7/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--kryaaa7/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "2355d7c5fc011d5fca3f9991a728be123d9fb7d2f20ea7e1aff7a22e3b60ad4b",
+   "video_sha256": "16cc0dc34ce70930e9fbe311a16f73bd9dba63c015040ff91c11446ee802f8ee",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--q4qt2kf.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--q4qt2kf.json
new file mode 100644
index 0000000000000000000000000000000000000000..d59197149e0cfccfb03008f2470ea7d5891a9532
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--q4qt2kf.json
@@ -0,0 +1,215 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-privileged--q4qt2kf",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 2694,
+   "sim_steps": 2694,
+   "horizon": 6000,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 3,
+   "stages_total": 3,
+   "fell": 0,
+   "replay_s": 21.95,
+   "bad_trajectory": 0,
+   "metric_rear_x_m": 14.301,
+   "metric_time_s": 26.94,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 111,
+   "agent_resets": 0,
+   "agent_replays": 13,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 5.425,
+   "peak_processes": 24,
+   "peak_disk_gb": 0.989,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-privileged--q4qt2kf/replay.mp4",
+  "duration_s": 26.96,
+  "fps": 25.0,
+  "frames": 674,
+  "sha256": "4da1dcf282f4cdf925db979369bc615235b0179dfc44ca659c94e6f4eea584e0"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 2,
+  "final": 2,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [
+     [
+      18.32,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 17.36,
+      "log_step": null,
+      "text_zh": "观察：视频 0–16.88 s，机器人沿跑道走到白色横杆前；视频 17.04–18.16 s，机器人蹬地、俯身前翻，两脚随身体转过横杆，到视频 18.32 s 身体已在坑侧并向坑底下降。越栏完成时间取视频约 18.32 s，事件此后保持。"
+     },
+     {
+      "t": 18.32,
+      "log_step": 152,
+      "text_zh": "观察：log step 152 的最终交付文件重生成校验输出为 success=True、2694 个控制步，hurdle 阶段在仿真 18.29 s 触发；骨盆、右脚、左脚向前过面高度分别为 1.300、1.432、1.470 m。log step 138–140 将该候选回放验证并选为交付，log step 151–152 确认保留且逐字节重生成；inputs/run.json 的评分字段 replay_success=1、deterministic=1、stages_reached=3，与视频翻越后落坑过程相符。"
+     },
+     {
+      "t": 17.6,
+      "log_step": null,
+      "text_zh": "推断：视频 17.28–17.68 s，机器人躯干贴近横杆并围绕该处从俯身转为倒置，说明翻越中横杆对躯干形成支撑；此前双脚在栏前地面支撑并蹬地。横杆与身体重叠、原始画面分辨率较低，不能确认左右拳垫是否触栏，不据此标双臂撑栏或抓握。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "机器人双脚先在栏前跑道支撑、蹬地，随后前翻越栏；视频 17.28–17.68 s 的贴栏转动支持躯干借横杆支撑的推断。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [
+     [
+      26.92,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 20.88,
+      "log_step": null,
+      "text_zh": "观察：视频 18.32–20.40 s，机器人落入蓝色坑近端、身体降到低处并转动，部分被横杆遮挡；视频 20.52–21.96 s 恢复直立和脚部支撑，视频 22.00–24.80 s 沿坑内斜坡向远岸行走，没有从坑侧绕行。"
+     },
+     {
+      "t": 26.92,
+      "log_step": null,
+      "text_zh": "观察：视频 25.28–25.40 s，机器人后脚离开坑远端边缘，随后在远岸红色跑道上减速；已密查视频 25.88–26.92 s，机器人保持直立、仅脚部接地，末帧仍在远岸。含连续站稳时长的节点完成记在视频末帧约 26.92 s，作为近似导航时间。"
+     },
+     {
+      "t": 26.92,
+      "log_step": 152,
+      "text_zh": "观察：log step 152 的最终交付重生成输出记录 out of the water jump 阶段在仿真 25.35 s、stand 和 success 在仿真 26.94 s，rear_x_m=14.301；inputs/run.json 的最终评分同样为 success=1、replay_success=1、metric_time_s=26.94、metric_rear_x_m=14.301。"
+     },
+     {
+      "t": 26.92,
+      "log_step": 152,
+      "text_zh": "推断：最终交付的成功裁判输出与视频出坑及末尾直立状态一致，确认全身出坑和连续 1.0 s 的仅脚部支撑、倾角及速度验收均通过；精确速度和倾角不能由固定远景独立测量。视频区间起点用已检查的末帧近似，精确完成值保留为评分器的仿真 26.94 s。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "机器人落入坑内后恢复脚部支撑；双脚依次在坑底、斜坡和远岸红色跑道承重行走，最终在远岸仅以脚部支撑站稳。横杆遮挡期间不能细分落坑瞬间各身体部位的接触。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "避开",
+      "note_zh": "视频 20.52–25.40 s，机器人在坑宽范围内恢复直立、沿坡前行至远岸，身体避开坑两侧红色跑道上表面。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better",
+   "value": 14.301
+  },
+  {
+   "name": "time_s",
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": 26.94
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人完成越过障碍栏、出坑并站稳两个节点，最终为 2/2；视频约 18.32 s 前翻清栏，落坑后恢复直立并沿坡走到远岸，视频末帧 26.92 s 保持站立，评分器在仿真 26.94 s 确认成功。最终交付经两次新进程回放成功且终态一致，原生 rear_x_m=14.301 m。无逐帧状态日志，远景及横杆遮挡使拳垫是否触栏、落坑瞬间的各部位接触和精确站稳阈值无法独立量化；阈值通过依最终交付的裁判输出确认，节点时刻按已检查视频帧近似标注。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标注与视频、评分字段一致：越栏约18.3 s，末帧26.9 s机器人直立于远岸，2/2成功，无失败层。未发现需改动的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:50:24Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--q4qt2kf/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--q4qt2kf/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "b73c484e77fafe3a67e820cac14b4e42df3d494bd9e3dfcf5e9e5a1bc3b2ce72",
+   "video_sha256": "4da1dcf282f4cdf925db979369bc615235b0179dfc44ca659c94e6f4eea584e0",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--yl68hst.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--yl68hst.json
new file mode 100644
index 0000000000000000000000000000000000000000..a8083e71c2a40c585602c0c7de905edcd31ef601
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-privileged--yl68hst.json
@@ -0,0 +1,219 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-privileged--yl68hst",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.3333,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 2650,
+   "sim_steps": 2650,
+   "horizon": 6000,
+   "episode_over": 0,
+   "progress": 0.3333,
+   "stages_reached": 1,
+   "stages_total": 3,
+   "fell": 0,
+   "replay_s": 12.39,
+   "bad_trajectory": 0,
+   "metric_rear_x_m": 11.65,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 249,
+   "agent_resets": 14,
+   "agent_replays": 5,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.325,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.853,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-privileged--yl68hst/replay.mp4",
+  "duration_s": 26.52,
+  "fps": 25.0,
+  "frames": 663,
+  "sha256": "85d8659abcc09b4b4cda8bccc5d871bc59a6e0c4ba41e69179ec2284fc416a63"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [
+     [
+      25.4,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 24.4,
+      "log_step": null,
+      "text_zh": "观察：视频0.0–24.0 s机器人在红色跑道上行走到白色横杆前并停留，视频24.2–24.8 s蹲下、蹬地并将双腿收向前方，身体旋转越过横杆。"
+     },
+     {
+      "t": 25.4,
+      "log_step": 381,
+      "text_zh": "观察：视频25.28–25.44 s机器人全身移至横杆后的蓝色坑内，越栏完成时间取约视频25.4 s；log step 381是已保存交付文件的验证回放，裁判输出stages=[hurdle]，并记录骨盆和两脚过栏高度分别为1.178、1.842、1.892 m，支持合规越栏。评分器字段stages_reached=1且deterministic=1与画面一致。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.0–24.4 s机器人双脚在栏前红色跑道交替承重行进，随后蹲下蹬地起跳越栏；未据会话中其他撑栏试验追加手部接触。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 25.6,
+      "log_step": null,
+      "text_zh": "观察：视频25.4–25.8 s机器人落入蓝色坑并翻倒，躯干降到坑底、双腿抬起后落下；视频26.0、26.2、26.4及26.48 s仍横卧在坑内，没有起身或到达远岸红色跑道。"
+     },
+     {
+      "t": 26.48,
+      "log_step": 381,
+      "text_zh": "观察：评分器字段metric_rear_x_m=11.65；log step 381交付文件验证回放只有hurdle阶段，未记出坑或站稳阶段。推断：结合视频末段全身仍在坑内，rear_x_m未超过13.66 m，出坑并站稳未达成；这不是仅凭总分为零作出的判断。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频25.6–26.48 s机器人在蓝色坑中翻倒，躯干和腿部落在坑底，由身体接地承重，未形成远岸仅脚部支撑的站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "value": 11.65,
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better"
+  },
+  {
+   "name": "time_s",
+   "value": null,
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "pit_crossed_standing",
+   "t": 25.6,
+   "log_step": null,
+   "where": "操作中",
+   "evidence_zh": "观察：视频25.4 s机器人已越到横杆后的蓝色坑内，视频25.6–25.8 s落地翻倒，至视频26.48 s结束仍未恢复站立、未向远岸出坑；此前越栏节点已完成，因此失败步骤为出坑并站稳。"
+  },
+  "unmet": {
+   "hurdle_cleared": true,
+   "pit_crossed_standing": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 25.6,
+    "log_step": null,
+    "evidence_zh": "观察：视频25.52–25.76 s机器人落入坑内时身体翻转，躯干倒下、两腿向上，随后身体横卧于坑底；视频26.0–26.48 s仍未恢复由脚承重的直立姿态。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：log step 362展示的交付生成脚本在双腿前收的35个控制步后令q=q2.copy()，随后150个控制步持续使用同一关节目标及零目标关节速度，没有转入落地伸腿、躯干稳定或坑内行走控制；log step 383–385重新执行该生成脚本，log step 381验证的交付终态也与评分记录一致。结合视频24.8–25.8 s持续翻转并倒在坑底、视频26.48 s仍未起身，支持越栏后的姿态与落地控制未能恢复站立；现有证据不足以进一步确定控制子类。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "无逐帧状态日志；固定全景分辨率不足以辨认每个瞬时碰撞部位、接触力及姿态跟踪误差，因此未进一步细分控制根因，也未将会话中的撑栏试验接触当成交付事实。评分器汇总fell=0，但视频25.6–26.48 s明确显示机器人翻倒并横卧坑内；平衡现象按视频判断，该汇总字段不用于否认可见翻倒。原生成功完成时间time_s未报告，保持null。"
+ },
+ "summary_zh": "观察：交付轨迹约在视频25.4 s完成越栏，最终达到两个DAG节点中的1个；评分器只记录hurdle阶段，rear_x_m=11.65，未达到13.66 m的出坑阈值。观察：机器人在视频25.6 s附近落入坑内翻倒，至视频26.48 s仍未起身，断在出坑并站稳；推断：生成脚本在收腿后持续保持固定关节目标，未完成落地及站立恢复，根因标为控制。观察：无逐帧状态日志，精细接触与控制误差无法核实；汇总fell=0与画面中的翻倒不一致，采用视频判断平衡现象。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "越栏约视频25.3 s完成、随后落入坑内翻倒、至结尾未出坑，节点值、断点、平衡现象与控制根因均与画面和评分字段一致。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:46:43Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--yl68hst/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-privileged--yl68hst/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "e9b1144a07c014094e74784c5deba2cbee32fca8f40da265baec32c8d93b5c31",
+   "video_sha256": "85d8659abcc09b4b4cda8bccc5d871bc59a6e0c4ba41e69179ec2284fc416a63",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--dshhjgo.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--dshhjgo.json
new file mode 100644
index 0000000000000000000000000000000000000000..14509523677ec788381bc3d73885ceda4573887c
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--dshhjgo.json
@@ -0,0 +1,216 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-standard--dshhjgo",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 3665,
+   "sim_steps": 15537,
+   "calls": 1000,
+   "rows": 13663,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 7098.1,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 3,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 6.142,
+   "peak_processes": 23,
+   "peak_disk_gb": 1.016,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-standard--dshhjgo/replay.mp4",
+  "duration_s": 36.68,
+  "fps": 25.0,
+  "frames": 917,
+  "sha256": "cf49591880e925e9ff34aff2dc4c7c4e0be5e7d23a4c1f4c06d45bf1a712e919"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 20.4,
+      "log_step": 247,
+      "text_zh": "观察：本次视频 19.6–20.6 s，机器人双脚仍在白色横杆前的红色跑道，双臂伸向横杆并调整姿态；视频 21.2–21.6 s 身体向后翻倒，未出现骨盆及两脚越到坑侧、全身清栏的过程。"
+     },
+     {
+      "t": 36.64,
+      "log_step": 291,
+      "text_zh": "观察：本次视频 22.0–36.64 s，机器人在横杆前地面翻滚、屈伸尝试起身，末帧仍倒在近岸；log step 290–291 的机器人服务裁判结果为 episode 6、control_steps=3665、sim_time_s=36.65、did not go over the hurdle、stages_reached=0/3，与画面一致。log step 238 是最后一次 reset，此前回合的动作不计入本次视频。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频开头至视频 21.2 s，机器人双脚接触红色跑道，行走接近横杆并在栏前站立调整；视频约 21.6 s 后躯干及腿部落在栏前跑道上，随后的起身动作也未离开近岸。"
+     },
+     {
+      "who": "双臂协作",
+      "how": "压/擦",
+      "note_zh": "观察：视频 19.6–20.6 s，机器人两臂同时前伸，固定拳垫在白色横杆近侧接触、抵压并调整；未见单手包围或夹住横杆，也未见双臂稳定承托全身越栏，故按实际抵压接触记录。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 36.64,
+      "log_step": 291,
+      "text_zh": "观察：本次视频从开始至视频 36.64 s，机器人未越过横杆进入蓝色坑后继续到达远岸；视频 35.6–36.64 s 仍倒在栏前近岸，远端红色跑道无人。log step 291 的服务裁判记录原生阶段为 0/3，亦未报告出坑或站稳。"
+     },
+     {
+      "t": 36.64,
+      "log_step": null,
+      "text_zh": "观察：成功参考视频 17.2–18.6 s 可见机器人身体翻过横杆进入坑侧，参考视频 26.92 s 可见机器人已在蓝色坑远端的红色跑道直立；本次同一全景视角中始终未出现这些位置变化。参考仅用于辨认达成状态，不作为本次进度证据。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "value": null,
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better"
+  },
+  {
+   "name": "time_s",
+   "value": null,
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "hurdle_cleared",
+   "t": 21.4,
+   "log_step": 257,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 20.4–21.2 s，机器人已在栏前抵压横杆、调整手臂和腿部，随后在视频约 21.4 s 向后翻倒，视频 21.6 s 躯干落在近岸跑道；视频至 36.64 s 未完成越栏。log step 257–258 记录最后回合的 recovery_fold、recovery_settle，log step 291 的服务裁判确认未越栏。"
+  },
+  "unmet": {
+   "hurdle_cleared": false,
+   "pit_crossed_standing": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 21.4,
+    "log_step": 257,
+    "evidence_zh": "观察：视频 21.2 s 机器人躯干明显向后倾，视频 21.4 s 身体接近横躺，视频 21.6 s 背部/躯干落到栏前跑道；此后多次屈伸和翻滚，视频 36.64 s 仍未恢复站立并完成越栏。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 21.4,
+    "log_step": 245,
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "推断：log step 244–245 显示 agent 在栏前从服务 stand 模式切换到自编 support_control.py，通过 lowcmd 输出基于双脚支撑假设的重力补偿和踝关节平衡力矩；log step 250、252、255 又连续修改踝关节增益、手臂和髋部目标。对应视频 20.4–21.6 s，机器人后倾加重并翻倒，说明该栏前全身控制未能维持稳定支撑；log step 257–289 的后续关节恢复序列也未使机器人站起并继续越栏。证据支持控制失败，但不足以唯一确定接触承载布局、姿态反馈或具体参数中的哪一项是根因。"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志；固定远景中的拳垫与横杆接触区域较小，无法量化接触力或区分两手的承载份额，也不能从画面精确重建重心与足底支撑关系。日志保留了控制代码和部分姿态读数，可支持控制层归因，尚不足以细分控制子类。评分器 reward 未报告 rear_x_m 或成功完成时间 time_s，两个原生指标均留空；本次失败的仿真终止时间不是成功完成时间。"
+ },
+ "summary_zh": "本次评分视频对应 log step 238 重置后的第6回合：机器人走到横杆前并用双臂抵压、调整，两个任务节点均未达成（0/2）。视频约 21.4 s，机器人在越栏准备动作中向后翻倒，后续恢复未能完成越栏，服务裁判在 log step 291 确认为未越栏、原生阶段 0/3。证据支持全身平衡现象及控制根因；无逐帧接触状态，无法进一步区分支撑接触设计与反馈控制问题。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "两节点均未达成、断点 hurdle_cleared（操作中）、全身平衡现象与控制根因均与视频和 log 一致；未发现需修改之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:59:04Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--dshhjgo/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--dshhjgo/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "8ae408fc9cd67e2156a0b01504d8ae75f1a18095e29cd1d045cc879494e7877f",
+   "video_sha256": "cf49591880e925e9ff34aff2dc4c7c4e0be5e7d23a4c1f4c06d45bf1a712e919",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--e3rwxda.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--e3rwxda.json
new file mode 100644
index 0000000000000000000000000000000000000000..be5140d1839aff0e33c1615226b7140ed622c7b9
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--e3rwxda.json
@@ -0,0 +1,204 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-standard--e3rwxda",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 200,
+   "sim_steps": 39921,
+   "calls": 2444,
+   "rows": 22050,
+   "resets": 29,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 7565.4,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.782,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.325,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-standard--e3rwxda/replay.mp4",
+  "duration_s": 2.04,
+  "fps": 25.0,
+  "frames": 51,
+  "sha256": "ecdc7cff9a2a958610768ff28ac219ee07f1d5c49b93073b79d2a501dd3b29bb"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.48,
+      "log_step": null,
+      "text_zh": "观察：视频0.00–0.64 s，机器人在白色障碍栏远前方的红色跑道上抬臂、伸腿并短促跃起，身体未到达横杆所在位置。"
+     },
+     {
+      "t": 1.16,
+      "log_step": null,
+      "text_zh": "观察：视频0.76–1.16 s，机器人落至双膝、身体前倾并扑倒；视频1.32–2.00 s持续趴在栏前跑道，骨盆和双脚均未越栏。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.00–0.64 s，机器人双脚在栏前红色跑道承重、蹬地；视频约0.76 s后双膝着地，随后躯干落在跑道上。未与横杆建立接触。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.0,
+      "log_step": null,
+      "text_zh": "观察：视频0.00–2.00 s的固定全景中，机器人始终位于横杆前方，未进入蓝色坑，也未到达坑远端红色跑道；末帧仍倒地。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "time_s",
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "hurdle_cleared",
+   "t": 0.8,
+   "log_step": null,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约0.48–0.64 s，机器人在远离障碍栏的跑道上短促跃起；视频约0.76–0.96 s落至双膝并持续向前倾倒，视频1.16 s已趴倒，至视频2.00 s未起身或越栏。这是首个未修复的跳跃步骤。"
+  },
+  "unmet": {
+   "hurdle_cleared": false,
+   "pit_crossed_standing": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 0.8,
+    "log_step": null,
+    "evidence_zh": "观察：视频0.76–1.16 s，机器人落地后双膝着地、躯干向前倒下，随后平趴在红色跑道；视频1.32–2.00 s一直未恢复脚部站立支撑。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 0.8,
+    "log_step": null,
+    "source_zh": "视频",
+    "evidence_zh": "推断：视频0.48–1.16 s，机器人短促起跳后未能维持落地姿态和脚部支撑，转为跪地并扑倒；全程远离横杆和水坑，未见外部障碍接触造成倒地，因此支持全身运动控制未能稳定完成起落，尚不能区分具体控制子类。"
+   }
+  ],
+  "gap_zh": "无逐帧状态日志，无法核验关节力矩、接触力或定位具体控制失稳机制。日志包含29次重置；log step 1015记录的是含助跑和后续行进的较长会话回合，不能套用于本段视频；log step 1026仅返回后台控制程序已启动，log step 1027为agent自述仍在等待，缺少随后各回合的完整反馈，无法将本段评分录像唯一对应到某次调用，也无法确定为何最终只保留这段短回合。节点和断点均以实际视频为准，不据此推定交付操作出错。评分器未报告rear_x_m或成功完成时间，原生指标值保留null。"
+ },
+ "summary_zh": "评分录像中，机器人在栏前远处短促跃起后，约视频0.8 s落至双膝并前扑，至视频2.00 s仍倒地；越过障碍栏和出坑并站稳均未完成，最终0/2。断点为越过障碍栏，现象是全身平衡丧失，视频支持一般控制失稳。日志含多次重置且最终后台执行缺少完整回合反馈，无法唯一对齐具体调用或进一步细分控制根因。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频0–2.04 s固定全景：机器人在栏前远处跃起、约0.8 s落地倒下并趴到末帧，未接近横杆或水坑；两节点false、断点hurdle_cleared、平衡+控制均与帧一致，未发现需改内容。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:49:49Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--e3rwxda/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--e3rwxda/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "eb52122c3267477dda7c04cc0e19146ce27cbd84a8a7ea61e7e718b0d400b732",
+   "video_sha256": "ecdc7cff9a2a958610768ff28ac219ee07f1d5c49b93073b79d2a501dd3b29bb",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--eakdsxp.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--eakdsxp.json
new file mode 100644
index 0000000000000000000000000000000000000000..34dd17044503e03349d738e3683a2fe2d203b247
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--eakdsxp.json
@@ -0,0 +1,224 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-standard--eakdsxp",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 2421,
+   "sim_steps": 6181,
+   "calls": 157,
+   "rows": 3423,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2268.8,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 3,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.415,
+   "peak_processes": 10,
+   "peak_disk_gb": 1.323,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-standard--eakdsxp/replay.mp4",
+  "duration_s": 24.24,
+  "fps": 25.0,
+  "frames": 606,
+  "sha256": "8d1acb4bbb9ef02b94c30e0564e26194f4a0073023d0a60aad6b64d8b1f0e0ac"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 16.0,
+      "log_step": 135,
+      "text_zh": "观察：视频 14.0–16.0 s，机器人沿红色跑道走到白色横杆近侧；视频 16.0–21.6 s，身体抵在横杆前，双腿继续迈动而整体位置基本不前进，未见骨盆及双脚越过横杆。"
+     },
+     {
+      "t": 22.4,
+      "log_step": 135,
+      "text_zh": "观察：视频 21.8–23.4 s，机器人摆臂、屈腿后向栏前跑道翻倒；视频 23.4–24.20 s，机器人躺在横杆近侧，未完成全身清栏。log step 137 的机器人服务（实时仿真控制接口）报告最后回合 episode=6、control_steps=2421、outcome=\"did not go over the hurdle\"、stages_reached=0/3，与视频结果一致。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 1.0–21.6 s，机器人双脚在栏前红色跑道上交替承重、向横杆行进；视频 22.4–24.20 s，翻倒后躯干等部位接触栏前跑道。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "压/擦",
+      "note_zh": "观察：视频约 16.0–21.6 s，机器人髋腹附近抵住固定白色横杆近侧，腿继续迈动但身体不能向前通过；画面不足以细分具体碰撞链接。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 24.2,
+      "log_step": 137,
+      "text_zh": "观察：通看视频 0–24.20 s，机器人始终未越过横杆进入蓝色坑，视频 23.4–24.20 s 仍倒在横杆近侧；未到达坑远端红色跑道，也未在那里站稳。log step 137 的机器人服务报告最后回合原生阶段 0/3；越栏前置未达，出坑步骤未启动。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "value": null,
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better"
+  },
+  {
+   "name": "time_s",
+   "value": null,
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "hurdle_cleared",
+   "t": 16.0,
+   "log_step": 135,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 16.0 s，机器人在走近横杆时髋腹附近抵住横杆，随后到视频 21.6 s 都未能前进越栏；视频约 22.4–23.4 s，后续屈腿摆臂动作使机器人失衡倒回栏前跑道，至视频 24.20 s 未修复越栏步骤。log step 135 记录最后回合的行进与 lowcmd 跳跃命令，log step 137 服务判定未越栏。"
+  },
+  "unmet": {
+   "hurdle_cleared": false,
+   "pit_crossed_standing": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "被挡/碰撞",
+    "t": 16.0,
+    "log_step": 135,
+    "evidence_zh": "观察：视频 15.8–21.6 s，机器人身体贴抵固定横杆，双脚仍迈动但身体持续停在栏前；横杆阻断向前通过，未见抬脚及全身越到坑一侧。"
+   },
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 22.4,
+    "log_step": 135,
+    "evidence_zh": "观察：视频 21.8–22.8 s，机器人在栏前摆臂屈腿、躯干明显倾斜并失去直立支撑；视频 22.8–23.4 s 向后倒地，视频 23.4–24.20 s 仍躺在栏前跑道。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "认知",
+    "item": "感知",
+    "log_step": 135,
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "推断：agent 的视觉定位器把机器人自身过滤掉，定位失效后脚本继续用过期位置，导致错过预定起跳位置并走到横杆前受阻。依据是 log step 34 中 track.robot_xy 的 m[:, 200:] = False；log step 143 的投影计算显示机器人到 x≈5 m 已落在约第205列；log step 135 的最后回合输出从仿真 9.1 s 起位置读数为 None，却继续行进，结束接近时仍使用 x=4.41894，随后又执行14次微步才触发跳跃。视频 9–16 s 可见机器人持续前进到横杆，视频 16–21.6 s 已抵栏，实际可见位置与脚本持有的位置不一致。"
+   },
+   {
+    "cat": "控制",
+    "item": "控制",
+    "log_step": 135,
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "推断：最后回合的低层跳跃控制未能维持机器人姿态，切回行走控制也未恢复支撑。log step 125 所写、log step 133 复制并在 log step 135 实际执行的脚本，向机器人提交约1.43秒的 lowcmd 预生成跳跃序列，再切回 walk；log step 135 同回合本体读数中躯干竖直方向分量 upz 从仿真 21.60 s 的0.99降到仿真 22.24 s 的0.29、仿真 22.88 s 的0.13，切回 walk 后为0.0。视频 21.8–24.20 s 独立显示该动作后失衡倒地；无法进一步区分具体力矩、轨迹或接触因素。"
+   }
+  ],
+  "gap_zh": "无逐帧状态日志，固定全景中机器人较小，不能精确测量三点过栏高度、触栏具体链接或瞬时接触力；但未越栏、未出坑和跌倒均清楚可见。log step 161 中 agent 自述“walked straight into the pit”不作为事实：视频末帧显示机器人仍倒在栏前。log step 145、147 的修正发生在最后回合结束后，未计入本次进度；评分字段未报告 rear_x_m 或成功完成时间 time_s，两个原生指标留空。"
+ },
+ "summary_zh": "本次评分对应最后实时回合 episode 6，机器人走到横杆前后未能越过，两个节点最终均未达到（0/2），出坑步骤未启动。视频约 16.0 s 起身体抵栏受阻，视频约 22.4 s 的跳跃动作失衡，最终倒在栏前；日志支持视觉定位失效后沿用过期位置，以及跳跃控制未能维持姿态。无逐帧状态日志，具体触栏链接及低层控制失稳机理仍不能确定；agent 事后修改的脚本没有改变已结束的评分回合。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "两个节点均未达到、断点在 hurdle_cleared（视频约 16 s 抵栏、22 s 后倒地）与视频和服务读数（outcome=did not go over the hurdle，stages 0/3，sim 24.21 s≈视频 24.24 s）一致；时钟与现象无明显错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:58:32Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--eakdsxp/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--eakdsxp/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "5721461c6a22fc144c8d8547f963f986f827d69b64d4092fb6b9b15486906e07",
+   "video_sha256": "8d1acb4bbb9ef02b94c30e0564e26194f4a0073023d0a60aad6b64d8b1f0e0ac",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--ecgdpi3.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--ecgdpi3.json
new file mode 100644
index 0000000000000000000000000000000000000000..281b9ddb31be57db1399fc64b80318edcb6dd689
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--ecgdpi3.json
@@ -0,0 +1,187 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-standard--ecgdpi3",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 1500,
+   "sim_steps": 1500,
+   "calls": 9,
+   "rows": 300,
+   "resets": 0,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 747.3,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.326,
+   "peak_processes": 8,
+   "peak_disk_gb": 0.527,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-standard--ecgdpi3/replay.mp4",
+  "duration_s": 15.04,
+  "fps": 25.0,
+  "frames": 376,
+  "sha256": "d7037aa282676fae247f34b73fa941712a296f7f1c9128f1d6c477741b7a4bf0"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频0.00 s机器人在红色跑道上，位于白色横杆前方；视频5.00 s、10.00 s和13.00 s依次显示机器人步行接近横杆，未越到蓝色坑一侧。"
+     },
+     {
+      "t": 15.0,
+      "log_step": 27,
+      "text_zh": "观察：本次视频14.00–15.00 s逐帧中机器人保持直立，双脚仍在栏前跑道上，未出现骨盆及两脚越过横杆并全身清栏的结果；log step 27的脚本仅向机器人提交walk步行命令。越栏节点未达成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：本次视频0.00–15.00 s，机器人双脚交替接触红色跑道、支撑身体向横杆行进；未见起跳越栏或脚踩横杆。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 15.0,
+      "log_step": null,
+      "text_zh": "观察：本次视频0.00–15.00 s机器人始终位于横杆前方；视频14.00–15.00 s逐帧未见机器人进入蓝色坑，更未见到达坑右侧远岸并站稳。该后续步骤未启动，节点未达成。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "value": null,
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better"
+  },
+  {
+   "name": "time_s",
+   "value": null,
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "hurdle_cleared",
+   "t": null,
+   "log_step": 39,
+   "where": "操作中",
+   "evidence_zh": "观察：本次视频14.00–15.00 s机器人仍在向白色横杆接近，尚未完成越栏，也未出现可见摔倒；log step 27结束步行提交后，log step 28–38仅观察或分析，log step 39返回“This request was blocked by Gemini's filters”的供应商过滤提示。推断：实时尝试在越栏步骤的接近阶段被外因截断，不能定位一个机器人自身出错的视频时刻，故t为空。"
+  },
+  "unmet": {
+   "hurdle_cleared": false,
+   "pit_crossed_standing": false
+  },
+  "phenomena": [],
+  "root_causes": [],
+  "gap_zh": "观察：inputs/有本次完整视频、运行日志及成功参考视频，无逐帧状态日志；run.json的reward记录success=0、resets=0、sim_steps=1500，log step 39为供应商过滤提示。视频末帧15.00 s仍显示机器人直立于栏前，未记录过滤解除后的继续动作或独立的物理失败。推断：按外因中断规则，本次保留目录未成功的结果及已观察进度，但应排除出机器人失败能力分析并重测；不把尚未越栏解释为步态失败或没尝试，现象和根因均留空。无状态日志也无法量化精确过面高度、接触力或rear_x_m；评分器未报告DAG中的rear_x_m及成功完成时间time_s，故两项原生指标均为null。"
+ },
+ "summary_zh": "观察：本次视频0.00–15.00 s机器人从起点步行至白色横杆前，两个节点均未达成；末秒逐帧中机器人保持直立，未进入蓝色坑。观察：log step 39返回供应商过滤提示，之前最后的动作提交仍是接近横杆的步行；推断：本次在越栏步骤进行中被外因截断，不作为机器人自身失败归因，现象和根因留空，宜重测。观察：无逐帧状态日志且无后续越栏动作，成功参考视频仅用于核对完成形态，不作为本次进度证据。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "视频末帧（15.0 s）机器人仍在白色横杆前方，两节点未达成与视频一致；log step 39 为供应商过滤提示，此前最后一次提交是 step 27 的步行，没有交付越栏动作，按外因截断将现象和根因留空并写 gap_zh 符合规则。未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:54:55Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--ecgdpi3/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--ecgdpi3/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "bd5c7f7e3e7defeddc2744796ca86e6cd3a7e67d5b7c3f76044ba70302db302f",
+   "video_sha256": "d7037aa282676fae247f34b73fa941712a296f7f1c9128f1d6c477741b7a4bf0",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--fjuscev.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--fjuscev.json
new file mode 100644
index 0000000000000000000000000000000000000000..5e7e133d7a4965a38c0db8e938ff8b3191de4bc8
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--fjuscev.json
@@ -0,0 +1,242 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-standard--fjuscev",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 6000,
+   "sim_steps": 52806,
+   "calls": 2246,
+   "rows": 24445,
+   "resets": 17,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 1387.1,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 4.497,
+   "peak_processes": 16,
+   "peak_disk_gb": 0.359,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-standard--fjuscev/replay.mp4",
+  "duration_s": 60.04,
+  "fps": 25.0,
+  "frames": 1501,
+  "sha256": "815a12205b96967987e1069627d65db6b694613758cfbbee8713f4f031be944a"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [
+     [
+      29.2,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 24.4,
+      "log_step": null,
+      "text_zh": "观察：视频23.6–26.8 s，机器人双脚仍在栏前，躯干/骨盆区域搭在白色横杆上，随后抬腿翻转；此时尚未全身清栏。"
+     },
+     {
+      "t": 29.2,
+      "log_step": 87,
+      "text_zh": "观察：视频28.36–29.32 s，机器人双腿上摆越过横杆，身体翻落至蓝色坑内；视频约29.2 s已无身体部分明显拖在栏前。log step 87执行最终控制脚本，服务裁判输出final time 1 / 3，log step 89核对该脚本仍为最终版本。推断：结合完整翻越画面和该回合首个原生阶段获认可，越栏节点已完成；首次全身清栏时刻约为视频29.2 s，误差约0.5 s，事件此后保留。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0–23.2 s机器人双脚交替接触栏前红色跑道推进；视频24.0–28.2 s躯干/骨盆区域搭压白色横杆并翻转，双腿随后抬起越栏。上肢的精确接触部位无法从小幅全景分辨，未据典型攀越方案推定双手撑栏。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 29.2,
+      "log_step": 87,
+      "text_zh": "观察：视频28.36–30.4 s机器人在越栏后头躯朝下翻落至坑内靠横杆处，没有形成脚部支撑的直立落地；log step 87的最终脚本在越栏动作后只调用观察等待，最终服务输出time、原生阶段1 / 3。"
+     },
+     {
+      "t": 60.0,
+      "log_step": 87,
+      "text_zh": "观察：视频30.4、31.0、35.0、45.0、59.0、59.8及60.0 s所检画面中无法辨认机器人，远岸红色跑道均没有机器人，坑内也没有可见的站立身影。成功参考视频24.0–26.8 s可清楚看到机器人由坑内走上远岸并保持直立，本次没有出现该可见结果。log step 87最终回合输出骨盆倾角约148.9°、原生阶段1 / 3；log step 88记录agent自述“lying prone in the pit”。推断：结合视频中翻落坑近端后被遮挡的过程与最终回合读数，机器人倒卧在坑近端的遮挡处，未出坑并站稳；倒卧位置和持续姿态并非后半段画面的直接观察。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频28.8–29.6 s机器人翻落至坑内靠横杆处，随后被遮挡；视频30.4、45.0、59.8及60.0 s无法辨认其身体或接触部位，远岸也没有机器人。log step 87最终回合输出骨盆倾角约148.9°、原生阶段1 / 3。推断：结合翻落过程与该读数，机器人由头躯等非脚部身体部位接触干燥坑底承托，未恢复脚部站立支撑；遮挡后的具体接触部位不可直接确认。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "value": null,
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better"
+  },
+  {
+   "name": "time_s",
+   "value": null,
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "pit_crossed_standing",
+   "t": 29.2,
+   "log_step": 87,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约29.2 s机器人虽已翻越横杆，却头躯朝下落入坑近端；视频30.4、31.0、35.0、45.0、59.0、59.8及60.0 s所检画面中无法辨认机器人，远岸没有机器人。log step 87的最终脚本仅保留越栏动作并等待，输出骨盆倾角约148.9°，服务裁判输出final time 1 / 3。推断：越栏步骤已完成，机器人随后倒卧在坑近端的遮挡处，未完成出坑并站稳；未修复的失败发生在该步骤的落坑及恢复阶段，故断点为该步骤。"
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 29.2,
+    "log_step": 87,
+    "evidence_zh": "观察：视频28.4–29.2 s机器人双腿向上、头躯向下翻入坑内；视频30.4、31.0、35.0、45.0、59.0、59.8及60.0 s所检画面中无法辨认机器人，远岸没有机器人，坑内也没有可见的站立身影。log step 87最终回合在越栏动作后打印骨盆倾角约148.9°，继而原生阶段停在1 / 3；log step 88记录agent自述“lying prone in the pit”。推断：机器人翻倒后倒卧于坑近端的遮挡处，未恢复以脚部支撑的直立姿态；后半段的持续倒卧是推断，不能由画面直接看见。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "log_step": 87,
+    "source_zh": "日志",
+    "evidence_zh": "观察：log step 74给出的控制代码用关节关键帧折髋、推脚并摆腿翻栏；log step 79将第二阶段改为直膝，最终log step 87重跑同一越栏序列后仅等待，没有起身或出坑控制。早先会话的log step 83、85尝试起身关键帧，反馈骨盆倾角仍约90–103°；agent在log step 86自述起身尝试只使机器人翻滚。推断：越栏后的全身姿态恢复控制未实现，落坑后无法恢复行进所需的脚部支撑，导致出坑步骤停滞；这些会话测试仅用于归因，不计入受评视频的动作或进度。证据不足以进一步判定为反馈、逆解或某个增益参数问题。"
+   }
+  ],
+  "gap_zh": "无逐帧状态日志；640×360固定全景中，翻落后的机器人在视频30.4–60.0 s所检画面中无法辨认，受横杆及坑沿等遮挡；后半段倒卧于坑近端是结合翻落过程与log step 87读数的推断，具体姿态和接触部位不可直接观察，不能量化三点过栏高度、精确全身清栏时刻或双手接触。越栏完成由视频中的完整翻越与log step 87最终回合的服务原生阶段1 / 3共同支持，时间取视频约29.2 s；远岸全景始终可见，足以判定未出坑。reward未报告rear_x_m及成功完成时间time_s，故两项原生指标为null；控制子类证据不足。",
+  "unmet": {
+   "hurdle_cleared": true,
+   "pit_crossed_standing": false
+  }
+ },
+ "summary_zh": "受评回合在视频约29.2 s完成翻栏，两个语义节点最终完成1/2；结合翻落画面和log step 87骨盆倾角约148.9°推断，机器人随后倒卧于坑近端的遮挡处，至视频60.0 s仍未出坑并站稳。断点为出坑并站稳，现象为全身失衡；最终控制脚本缺少有效的落坑后姿态恢复，早先会话的起身关键帧也未能使机器人直立，支持控制层归因。无逐帧状态日志，后半段机器人不可辨认，精确过栏阈值和上肢接触不可量化，但最终回合裁判的原生阶段1/3与视频结果相符。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 7,
+    "summary_zh": "节点值、时间（越栏约29.2 s）、断点（出坑并站稳）与现象（平衡）均与视频及log step 87一致；仅有一处把「看不见」写成「看见」的观察表述需要改为推断。",
+    "n_issues": 1,
+    "n_major": 0
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点（越栏约视频29.2 s完成、出坑并站稳未完成）、断点、全身·平衡现象及控制根因与视频和log step 87一致；上一轮指出的观察/推断混写已修正，未发现新的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:00:51Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--fjuscev/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--fjuscev/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--fjuscev/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--fjuscev/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "4fb706fef6254a1c809c1ec2a2827a580dfdeb829e753c33abb6c8edb656e3fb",
+   "video_sha256": "815a12205b96967987e1069627d65db6b694613758cfbbee8713f4f031be944a",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--hd7munb.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--hd7munb.json
new file mode 100644
index 0000000000000000000000000000000000000000..a5bda031fec84d2a94153153fb669ee482732e40
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--hd7munb.json
@@ -0,0 +1,202 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-standard--hd7munb",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 1668,
+   "sim_steps": 21081,
+   "calls": 191,
+   "rows": 2336,
+   "resets": 12,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 7171.1,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.978,
+   "peak_processes": 13,
+   "peak_disk_gb": 1.302,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-standard--hd7munb/replay.mp4",
+  "duration_s": 16.72,
+  "fps": 25.0,
+  "frames": 418,
+  "sha256": "526564b6782e02371d9ef5a08d4ea317438cb5d17f51a65bf8388665edafc946"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 14.8,
+      "log_step": 272,
+      "text_zh": "观察：视频 0–14.8 s，机器人沿红色跑道走近白色固定横杆；视频 15.6–16.4 s 仍在横杆前方，双臂向栏伸出。log step 272 是最后一次 reset 后的接近、站立和双腿同步抬升脚本；此前各局不计入本次进度。"
+     },
+     {
+      "t": 16.68,
+      "log_step": 274,
+      "text_zh": "观察：视频 16.44–16.68 s，机器人在栏前屈腿、躯干前倾并下沉，两脚没有越到蓝色坑一侧，全身未清栏；最后一帧也未显示越栏完成。成功参考视频约 17–19 s 可见机器人整体从栏前翻越到坑内，本次没有出现这一可见结果。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0–16.4 s，机器人两脚交替接触红色跑道，支撑接近横杆并停步；随后在原处屈腿抬升尝试越栏。双手是否与横杆实际承重接触无法由此远景确认，不据伸臂动作推定撑栏。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 16.68,
+      "log_step": null,
+      "text_zh": "观察：视频 0–16.68 s，机器人始终留在白色横杆的接近侧，未进入蓝色坑，更未到达坑远端红色跑道；该全景可见出坑目标区域。成功参考视频约 24–26.9 s 的到达远岸结果在本次没有出现，且前置越栏未完成。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "value": null,
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better"
+  },
+  {
+   "name": "time_s",
+   "value": null,
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "hurdle_cleared",
+   "t": 16.5,
+   "log_step": 272,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 16.44–16.68 s，机器人在栏前开始屈腿抬升，躯干随即前倾下沉，未把骨盆和两脚越过横杆，直至片尾未修复。log step 272 的最后一局脚本同步改变左右髋、膝关节目标；此局之后无新的动作提交记录。"
+  },
+  "unmet": {
+   "hurdle_cleared": false,
+   "pit_crossed_standing": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 16.5,
+    "log_step": 272,
+    "evidence_zh": "观察：视频 16.4 s 机器人尚在栏前直立，视频 16.48–16.68 s 抬腿时躯干快速向横杆方向倾斜、骨盆下沉，站立支撑失稳；片尾没有恢复直立，也没有完成越栏。画面止于前倾下沉阶段，未显示完整摔倒着地过程。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：log step 272 的最后一局脚本在停步后用 28 个 10 ms 的 lowcmd 行同步驱动双腿屈髋屈膝，整段使用动作前读取的骨盆姿态计算补偿，没有在这段抬升中按新姿态更新支撑控制；视频 16.44–16.68 s 对应出现快速前倾下沉而未清栏，支持越栏动作的全身控制未维持稳定。现有资料不足以细分为接触承载布局、轨迹跟踪或增益参数问题。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "无逐帧状态日志，固定远景中的机器人较小，不能确认两只拳垫与横杆的接触及承重情况，也不能精确恢复接触力或各关节跟踪误差，因此根因只标到控制，不细分。视频结束于前倾下沉阶段，不能推断其后是否摔倒或可能恢复；评分器 reward 未报告 rear_x_m 和 time_s，原生指标保留 null。日志包含多次 reset 的会话尝试，本次节点与现象只依据最后一局视频，未使用早先各局的违规或跌倒作为本局事实。"
+ },
+ "summary_zh": "本次评分局中机器人沿跑道接近横杆，在视频约 16.5 s 的双腿抬升尝试中前倾下沉，未完成越栏，也未开始穿坑到远岸，最终为 0/2 节点。最后一局动作脚本与片尾画面支持全身控制失稳；无逐帧状态日志，手部是否承重撑栏及具体控制子因无法确认。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点 0/2、断点 hurdle_cleared（操作中）、平衡现象与控制根因均与视频末段（16.2–16.68 s 栏前屈腿前倾、未越栏）和 log step 272（最后一局 reset 后接近并以 28 行 lowcmd 同步抬腿）一致；未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:00:48Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--hd7munb/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--hd7munb/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "16f5bc7df5888e58109fe5a1f037fa3b728eef54dc6c67ea2e6527aa9ed8f1a1",
+   "video_sha256": "526564b6782e02371d9ef5a08d4ea317438cb5d17f51a65bf8388665edafc946",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--rxbbr6c.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--rxbbr6c.json
new file mode 100644
index 0000000000000000000000000000000000000000..19506192fc7a37f33843a6b45e40361adff210e0
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--rxbbr6c.json
@@ -0,0 +1,214 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-standard--rxbbr6c",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 6000,
+   "sim_steps": 14915,
+   "calls": 234,
+   "rows": 7253,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 1712.9,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 3,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.581,
+   "peak_processes": 11,
+   "peak_disk_gb": 0.346,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-standard--rxbbr6c/replay.mp4",
+  "duration_s": 60.04,
+  "fps": 25.0,
+  "frames": 1501,
+  "sha256": "60f1066101c5e6aa7c38081f0d5f11935cbdd02561eb255361b6ff3a2f85802a"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 17.0,
+      "log_step": 129,
+      "text_zh": "观察：视频0–17.0 s中机器人以双脚支撑走近白色横杆，视频17.0–19.6 s仍在横杆近侧摆动双臂和调整身体，未见骨盆及两脚整体越到蓝色坑一侧。log step 129记录该段来自最后重置后的第6回合，三组手臂姿态的holding读数均为(0,0)。"
+     },
+     {
+      "t": 20.4,
+      "log_step": 130,
+      "text_zh": "观察：视频19.8–20.6 s中机器人在横杆近侧倾斜并倒地；视频21.0–40.0 s反复屈腿、摆臂和贴地移动，未完成越栏，视频59.12、59.6及60.0 s仍倒在栏前跑道。"
+     },
+     {
+      "t": 60.0,
+      "log_step": 135,
+      "text_zh": "观察：log step 135中机器人服务报告第6回合control_steps=6000、done=True、outcome=time、stages_reached=0/3；与视频未越栏的结果一致。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0–19.6 s双脚交替接触红色跑道，承重接近障碍栏；视频20.4 s以后腿及躯干倒在同一侧跑道上，恢复动作仍以地面为支承，未见脚踩上横杆。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 25.0,
+      "log_step": 130,
+      "text_zh": "观察：视频20.6–30.0 s中机器人始终倒在白色横杆近侧红色跑道，蓝色坑内及远岸均没有机器人；恢复动作没有使其进入坑内。"
+     },
+     {
+      "t": 60.0,
+      "log_step": 135,
+      "text_zh": "观察：视频32.0–60.0 s机器人继续在栏前地面活动后停住，视频59.12、59.6及60.0 s均未在坑远岸站立；log step 135的服务最终阶段完成数为0/3。出坑并站稳从未达成，越栏前置也未完成。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "value": null,
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better"
+  },
+  {
+   "name": "time_s",
+   "value": null,
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "hurdle_cleared",
+   "t": 20.2,
+   "log_step": 129,
+   "where": "操作中",
+   "evidence_zh": "观察：视频17.0–19.6 s机器人在栏前伸臂和调整姿态，视频19.8–20.6 s逐渐倾斜并倒在近侧跑道；随后直到视频60.0 s都未完成越栏。log step 129记录最后回合的撑栏姿态尝试，holding均为0，第二、第三组姿态后IMU俯仰读数约为−87°、−90°。"
+  },
+  "unmet": {
+   "hurdle_cleared": false,
+   "pit_crossed_standing": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 20.2,
+    "log_step": 129,
+    "evidence_zh": "观察：视频19.8–20.6 s机器人由双脚站立变为身体倾倒贴地；视频22–40 s反复屈腿、摆臂仍无法恢复脚部独立承重，视频59.12–60.0 s仍倒地。log step 130的恢复动作结束后俯仰读数再次达到−90°。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：栏前全身姿态控制未维持稳定支撑。log step 126将撑栏尝试设为三组固定躯干及肩部关节目标，log step 129记录第一组后俯仰约−8°，第二组后约−87°，且双手holding一直为0；视频19.6–20.6 s对应出现倾倒。log step 130改用四组固定关节目标起身，读数由−65°回到−90°，视频22–26 s也未站起。证据支持控制失效，但不足以区分接触布局与反馈跟踪的具体责任。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "无逐帧状态日志，固定全景中的机器人较小，无法确认拳垫与横杆是否有瞬时轻触，故不记已建立手撑栏或抓握。精确接触力、支撑分布及失稳的控制子类型不可确定；reward未报告rear_x_m或成功完成时间，原生指标均保留null。log step 129的x=10.26、y=2.27来自agent图像定位估计，log step 142据此声称已过栏平面；视频仍显示机器人在横杆近侧倒地，节点以视频及服务阶段结果为准。"
+ },
+ "summary_zh": "最终计分的第6回合完成0/2个DAG节点：机器人走到栏前后，在视频约20.2 s尝试撑栏姿态时失衡倒地，后续恢复未成功，始终未越栏或到达坑远岸。服务最终报告time、原生阶段0/3；根因证据支持全身姿态控制失效，不能细分其控制子类型。无状态日志，短暂手栏接触及精确几何量无法确认，agent的位置估计未用于确认越栏。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频显示机器人走到栏前，约19.6–20.4 s倾倒，之后一直倒在栏前，未越栏也未到坑远岸；两个节点false、breakpoint在hurdle_cleared、平衡现象及控制根因均有视频和服务读数（0/3阶段）支持，未发现需要修改之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:58:38Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--rxbbr6c/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--rxbbr6c/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "f9a7daf38e65ed46930b299156812fe11f4af8b6eda43148bc81a12ab9bcaef1",
+   "video_sha256": "60f1066101c5e6aa7c38081f0d5f11935cbdd02561eb255361b6ff3a2f85802a",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--znaszun.json b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--znaszun.json
new file mode 100644
index 0000000000000000000000000000000000000000..20306192e649c9d87d0adf9fe0d608d53cc83688
--- /dev/null
+++ b/phase2/embodied-olympics-fun/steeplechase/trajectories/embodied-olympics-fun-steeplechase-i00-standard--znaszun.json
@@ -0,0 +1,219 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-steeplechase-i00-standard--znaszun",
+ "taskKey": "embodied-olympics-fun/steeplechase",
+ "dag_sha256": "d1654f75d2327bdc38995638268d9415b735f4a2b26d39708a7baeb8fa7d5b71",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 2808,
+   "sim_steps": 21099,
+   "calls": 532,
+   "rows": 10630,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2132.4,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 3,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.17,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.241,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-steeplechase-i00-standard--znaszun/replay.mp4",
+  "duration_s": 28.12,
+  "fps": 25.0,
+  "frames": 703,
+  "sha256": "f6cf26e95f03dbccd8c5f63fda6bf60098ff6e3a7b0b9ef67b4ad9a5a5b4f4d0"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 2,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "hurdle_cleared": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 20.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0–20.0 s，机器人沿红色跑道走向白色障碍栏，停在栏前，两脚仍在近岸，上肢前伸到横杆处。"
+     },
+     {
+      "t": 22.84,
+      "log_step": null,
+      "text_zh": "观察：视频 22.60–23.04 s，机器人在栏前屈腿下沉，头部降至接近横杆高度；随后至视频 28.08 s 仍缩在栏前，未见骨盆及两脚越到坑侧或全身清栏。"
+     },
+     {
+      "t": 28.08,
+      "log_step": 56,
+      "text_zh": "观察：log step 50 的服务输出确认最后重置为 episode 6；log step 56 查询该回合裁判结果为 failed、did not go over the hurdle、stages 0/3，log step 57 确认 episode 6 已结束。此结果与视频末帧仍在栏前一致。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "机器人双脚在红色近岸跑道交替支撑完成接近，之后在栏前支撑身体；未见脚踏上横杆完成承重。"
+     },
+     {
+      "who": "手指/夹爪",
+      "how": "支撑",
+      "note_zh": "视频 20.0–22.6 s 可见上肢末端抵在白色横杆处尝试撑扶；结合 log step 52 的左右肩电机力矩及 log step 55 的 agent 自述，推断主要由左手固定拳垫压撑，右手是否实际承重无法确认，故不标双臂协作或抓握。"
+     }
+    ]
+   },
+   "pit_crossed_standing": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 28.08,
+      "log_step": 56,
+      "text_zh": "观察：视频全程至 28.08 s，机器人始终未从栏前进入蓝色坑并到达其远端红色跑道；末段仍低伏在栏前，远岸无人。log step 56 的最后回合服务结果为 stages 0/3。"
+     },
+     {
+      "t": 28.08,
+      "log_step": null,
+      "text_zh": "推断：固定全景足以显示全身到达远岸；已检查的成功参照视频 24.0–26.0 s 可见机器人从坑中走上远岸，本次全程没有这一位置变化，且越栏前置未达到，因此出坑并站稳未完成。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "rear_x_m",
+   "unit": "m",
+   "threshold": 13.66,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "time_s",
+   "unit": "s",
+   "threshold": null,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "hurdle_cleared",
+   "t": 22.8,
+   "log_step": null,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 22.60–23.04 s，机器人正于栏前撑扶横杆，身体突然屈腿下沉至低姿态，至视频 28.08 s 未恢复或越栏。log step 56 的最后回合服务裁判结果为 did not go over the hurdle、stages 0/3，因此断在越过障碍栏步骤。"
+  },
+  "unmet": {
+   "hurdle_cleared": false,
+   "pit_crossed_standing": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 22.8,
+    "log_step": null,
+    "evidence_zh": "观察：视频 22.60–23.04 s，机器人从栏前直立姿态快速塌落，躯干与头部降至横杆附近及以下，之后至视频 28.08 s 保持低伏姿态，没有恢复直立支撑。推断：这是越栏尝试中的支撑失稳；agent 也在 log step 59 自述最后回合在栏前摔倒，但其单腿抬起的具体叙述不作为画面已证实的动作。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "推断：越栏撑扶阶段的全身姿态控制未维持住支撑。log step 30 展示的 hold/move 实现把固定或线性插值的关节目标整段提交给 lowcmd，未在段内按躯干姿态调整目标；log step 52、55 的最后回合脚本沿用这些动作，肩部接触修正后继续关节目标序列，视频 22.60–23.04 s 则显示身体在栏前塌落。没有足够证据区分接触承载布局与姿态反馈的各自贡献，故保留在控制大类。"
+   }
+  ],
+  "gap_zh": "无逐帧状态日志，640×360 固定全景中机器人较小，不能精确测定三点过栏高度、各手承重或身体触地点。日志最后回合的 proprio time_s 与视频动作没有获得可靠对齐：log step 52 在其记录的回合 26.4 s 仍为 done=False，log step 55 叙述抬右腿后失败，而视频约 22.8 s 已出现下沉并保持至末帧；因此未把日志时间换算成视频时间，也未把单腿抬起或低位过面的精确机制当作视觉事实。rear_x_m 和成功完成时间均未由评分器字段报告。"
+ },
+ "summary_zh": "本次评分的是最后实时回合 episode 6：机器人走到栏前并尝试用上肢撑扶，但未越栏，出坑并站稳也未启动，最终完成 0/2 个节点。视频约 22.8 s 机器人在栏前支撑失稳并塌落，之后未恢复；日志中的关节目标执行方式支持控制失稳的归因。无状态日志且全景分辨率有限，日志动作时间与视频未可靠对齐，不能进一步确定单腿动作和三点过栏的精确失败机制。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "视频显示机器人在栏前撑扶后约22.8 s下沉并保持低伏至末帧，未越栏；log step 56服务结果为failed、stages 0/3，两节点false、断点hurdle_cleared、平衡现象与控制归因均有证据支持，未发现需改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:02:51Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--znaszun/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-steeplechase-i00-standard--znaszun/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "81b64e35e25737f0ab0598f7e77d7ed86e345bef8867058136e3f156d144e75c",
+   "video_sha256": "f6cf26e95f03dbccd8c5f63fda6bf60098ff6e3a7b0b9ef67b4ad9a5a5b4f4d0",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--2bzyfpf.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--2bzyfpf.json
new file mode 100644
index 0000000000000000000000000000000000000000..0fa885f2f2e11e4413fe5b996c8114f5f6d6c15f
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--2bzyfpf.json
@@ -0,0 +1,182 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-privileged--2bzyfpf",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 404,
+   "sim_steps": 404,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 5.73,
+   "bad_trajectory": 0,
+   "metric_contact_s": 0.026,
+   "metric_impulse_n_s": 7.68,
+   "metric_peak_force_n": 774.8,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 90,
+   "agent_resets": 4,
+   "agent_replays": 2,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.803,
+   "peak_processes": 8,
+   "peak_disk_gb": 0.254,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--2bzyfpf/replay.mp4",
+  "duration_s": 4.08,
+  "fps": 25.0,
+  "frames": 102,
+  "sha256": "90f16f47e01c009b3b1e0860f911e38f42b08436552872e74fa160eead0a75f2"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [
+     [
+      4.04,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.96,
+      "log_step": null,
+      "text_zh": "观察：视频 0.48–0.92 s，银色机器人屈膝抬腿并向假人蓝色护具伸脚；视频 0.92–0.96 s 足部到达护具下部，随后视频 1.00–1.36 s 屈膝收腿，另一脚始终在红色垫面支撑。"
+     },
+     {
+      "t": 4.04,
+      "log_step": null,
+      "text_zh": "观察：视频 1.44–2.04 s，机器人放下踢击腿并调整身体；视频 2.04–4.04 s 双脚在红色垫面上，躯干保持直立，未见手、膝或躯干撑地；末秒全部帧均保持站立姿态。"
+     },
+     {
+      "t": 4.04,
+      "log_step": 180,
+      "text_zh": "观察：log step 174 记录最终轨迹保存成功；log step 180 对该交付文件的独立重放返回 success=true、stages=[kick,stand]、executed_rows=404，kick 事件在仿真 0.984 s，stand 与 success 事件在仿真 4.04 s；inputs/run.json 的评分器字段也报告 replay_success=1、deterministic=1、stages_reached=2、metric_impulse_n_s=7.68、metric_contact_s=0.026。"
+     },
+     {
+      "t": 4.04,
+      "log_step": 180,
+      "text_zh": "推断：交付轨迹每控制步为 0.01 s（log step 158 的动作段步数与时长对应），log step 180 的 404 个控制步对应视频 4.04 s；仿真 0.984 s 的踢击事件与视频 0.92–1.00 s 的触靶及收脚相符，仿真 4.04 s 的完成事件与视频 4.04 s 末帧站立相符，故将完整踢击并站稳节点首次完成记为视频 4.04 s。接触合法性、冲量及连续站稳计数由交付重放评分器确认。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 0.92–0.96 s，抬起的脚向假人蓝色躯干护具短促踢击，随后收回；结合 log step 158 的交付脚本可辨为右脚。候选名称“击打”，现有接触方式无踢击对应项，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.48–1.36 s，踢击腿抬起、触靶及回收期间，另一只脚支撑红色垫面；结合 log step 158 可辨为左脚。随后右脚落地，视频 2.04–4.04 s 双脚支撑垫面完成站稳。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better",
+   "value": 7.68
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.026
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人短促踢中假人护具，收腿落地后完成连续站稳，唯一节点“踢中护具并站稳”在视频 4.04 s 达成，最终为1/1。交付文件的最终重放与评分器均确认成功，护具冲量为7.68 N·s、接触时长为0.026 s，均达标。没有逐帧状态日志，25 fps视频不能独立精确核验毫秒级接触、踝局部接触点或站稳数值阈值，这些判据采用该交付文件的评分器结果。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "标注与视频、log 和评分器读数一致：视频约 0.92–0.96 s 踢中护具，其后收腿落地站稳，节点在视频 4.04 s 完成。控制步到视频时间的换算（0.01 s/步，404 步对应 4.04 s）有依据，接触标注合理。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:44:05Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--2bzyfpf/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--2bzyfpf/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "0f595bd484a3c50fb940c2186b4500b375517d1069c899ec73d03fc6989d5d1f",
+   "video_sha256": "90f16f47e01c009b3b1e0860f911e38f42b08436552872e74fa160eead0a75f2",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--9pgabbe.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--9pgabbe.json
new file mode 100644
index 0000000000000000000000000000000000000000..e607b86561ae5c13e6f10923cb37502dd4500f17
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--9pgabbe.json
@@ -0,0 +1,202 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-privileged--9pgabbe",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 0,
+  "reward": 0.5,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 351,
+   "sim_steps": 351,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 0.5,
+   "stages_reached": 1,
+   "stages_total": 2,
+   "fell": 1,
+   "replay_s": 3.35,
+   "bad_trajectory": 0,
+   "metric_contact_s": 0.033,
+   "metric_impulse_n_s": 25.43,
+   "metric_peak_force_n": 1378.9,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 169,
+   "agent_resets": 71,
+   "agent_replays": 5,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.63,
+   "peak_processes": 11,
+   "peak_disk_gb": 0.297,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--9pgabbe/replay.mp4",
+  "duration_s": 3.52,
+  "fps": 25.0,
+  "frames": 88,
+  "sha256": "96871a9013e926f7a130c010b5a76a0931937b8ad04a35c8708954f470537d3a"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.8,
+      "log_step": null,
+      "text_zh": "观察：视频0.00–2.60 s机器人在假人前站立并调整支撑，视频2.68–3.12 s抬起右腿，左脚留在红色垫面支撑。"
+     },
+     {
+      "t": 3.2,
+      "log_step": 214,
+      "text_zh": "观察：视频3.16–3.20 s机器人右脚伸向并击中假人胸腹前的蓝色白圆标护具；log step 214对最终交付文件重新生成并正式回放，裁判仅记录kick阶段、冲量25.43 N·s、接触时长0.033 s，随后failed=fell。inputs/run.json的最终评分字段与此一致，踢击阶段通过。"
+     },
+     {
+      "t": 3.48,
+      "log_step": 190,
+      "text_zh": "观察：视频3.24–3.48 s机器人躯干持续后仰、骨盆下降，右腿仍在空中，未收腿形成双脚站稳；末帧视频3.48 s仍在倒下。log step 190对交付文件的正式回放仅有kick阶段且success=false。完整节点要求踢后连续双脚站稳，故本节点从未完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频3.16–3.20 s机器人右脚击中假人蓝色护具；候选名称“击打”，现有接触方式没有踢击对应值，依S16记其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频2.68–3.20 s机器人抬起右腿踢靶时，左脚在红色垫面上支撑身体；踢后未形成稳定双脚支撑。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better",
+   "value": 25.43
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.033
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "protector_kicked_and_stood",
+   "t": 3.24,
+   "log_step": 190,
+   "where": "操作中",
+   "evidence_zh": "观察：视频3.16–3.20 s机器人右脚击靶后，视频3.24 s躯干明显向后倾倒，至视频3.48 s骨盆持续降低、踢击腿仍悬空，没有完成收腿落地与连续站稳。log step 190确认交付回放仅达到kick阶段，因fell终止；断点属于“踢中护具并站稳”这一完整动作。"
+  },
+  "unmet": {
+   "protector_kicked_and_stood": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 3.24,
+    "log_step": 190,
+    "evidence_zh": "观察：视频3.24–3.48 s机器人踢靶后躯干快速后仰、双臂上扬、骨盆下沉，未恢复直立双脚支撑；log step 190的交付回放结果为failed=fell，最终评分fell=1。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "log_step": 204,
+    "evidence_zh": "推断：log step 204展示最终生成脚本逐行发送预先保存的lowcmd指令，未根据当前整身姿态在线调整恢复动作；log step 214再次生成与回放同一交付文件，仍仅达到kick并跌倒。结合视频3.24–3.48 s踢后后仰持续扩大、未及时收腿恢复支撑，说明该交付动作的踢后平衡恢复控制未奏效；证据不足以细分为逆解、反馈或某个参数错误。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志，无法精确核验接触点、关节跟踪误差和失衡力矩；冲量与接触时长取自最终评分和经核对的交付回放结果。日志中的较软踢击、冻结支撑脚测量等后续实验没有替换最终交付，不能把那些实验的诊断直接当作本次失衡的确定机制。"
+ },
+ "summary_zh": "机器人在视频约3.16–3.20 s用右脚击中护具，裁判确认有效踢击，但踢后未落地站稳，完整语义节点最终为0/1；原生progress=0.5仅表示kick阶段通过。视频约3.24 s起机器人持续后仰倒下，断在“踢中护具并站稳”的操作中，现象为平衡失败，交付动作的恢复控制未奏效。无逐帧状态日志，具体失衡机制及毫秒级接触只可分别保留为未细分归因与评分器数值。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点未完成（踢击阶段通过、踢后后仰倒下）、断点、平衡现象与控制根因均有视频和评分字段支持；未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:50:28Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--9pgabbe/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--9pgabbe/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "8aca0ecb577d58a86f67c9b5e6b0a76922ca5d28f9cdece2ace125656a2d9bac",
+   "video_sha256": "96871a9013e926f7a130c010b5a76a0931937b8ad04a35c8708954f470537d3a",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--kk2pkmp.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--kk2pkmp.json
new file mode 100644
index 0000000000000000000000000000000000000000..a0acdf06ddc750e48d8bf0aa9c9f827b0b4678e1
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--kk2pkmp.json
@@ -0,0 +1,201 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-privileged--kk2pkmp",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 0,
+  "reward": 0.5,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 1500,
+   "sim_steps": 216,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 0.5,
+   "stages_reached": 1,
+   "stages_total": 2,
+   "fell": 1,
+   "replay_s": 2.76,
+   "bad_trajectory": 0,
+   "metric_contact_s": 0.044,
+   "metric_impulse_n_s": 9.15,
+   "metric_peak_force_n": 349.5,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 2,
+   "agent_resets": 0,
+   "agent_replays": 8,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 5.359,
+   "peak_processes": 18,
+   "peak_disk_gb": 0.984,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--kk2pkmp/replay.mp4",
+  "duration_s": 2.2,
+  "fps": 25.0,
+  "frames": 55,
+  "sha256": "7e1255c5d5ef62956ec4467342b4cc3527207649ec1f1d672b6928dbca0df983"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.28,
+      "log_step": 247,
+      "text_zh": "观察：视频 1.00–1.24 s 机器人抬右腿，视频 1.28–1.32 s 右脚到达假人蓝色护具前表面，视频 1.36–1.48 s 收腿离靶。log step 247 对最终生成的交付文件回放，裁判返回 stages=[kick]、冲量9.15 N·s、接触时长0.044 s；run.json 的最终评分字段与这些指标一致，说明有效踢击阶段通过，但这不是完整节点完成。"
+     },
+     {
+      "t": 1.56,
+      "log_step": 249,
+      "text_zh": "观察：视频 1.52–1.76 s 机器人放低右腿时躯干明显后仰并向远离假人的方向退移；视频 1.80–2.16 s 后仰加剧、骨盆降低，末帧机器人倒向垫面，未出现踢后双脚连续站稳。log step 249 对同一交付文件再次回放返回 failed=fell、stages=[kick]、success=false，终态哈希与 log step 247 相同；最终评分为 fell=1、stages_reached=1/2。"
+     },
+     {
+      "t": 2.16,
+      "log_step": null,
+      "text_zh": "推断：有效踢击后收腿站稳的完整动作从未成立，因此本事件节点无达成区间、终值为false。成功参考视频 1.60–4.04 s 可见机器人收腿后恢复双脚站立并持续保持；该参考仅用于辨认完成外观，本次视频 1.60–2.16 s 呈持续失衡。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 1.28–1.32 s 机器人右脚击中假人的蓝色胸腹护具，随后收回；最终评分记录有效kick。候选名称“击打”，接触方式表没有踢击专项，故用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 1.08–1.48 s 机器人抬右腿踢击时左脚支撑红色垫面；视频 1.56–1.76 s 收腿后脚部在垫面上承重、调整落点，但未恢复稳定站立，随后向后倒下。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better",
+   "value": 9.15
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": 0.044
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "protector_kicked_and_stood",
+   "t": 1.56,
+   "log_step": 247,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 1.28–1.32 s 机器人踢中护具后收腿，视频约 1.56 s 躯干明显后仰，落脚及后退未能止住倾倒，至视频 2.16 s 倒向垫面，完整踢击及站稳动作未完成。log step 247 的最终交付回放与最终评分均显示有效kick之后因fell终止，未达到stand。"
+  },
+  "unmet": {
+   "protector_kicked_and_stood": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 1.56,
+    "log_step": 249,
+    "evidence_zh": "观察：视频 1.56–2.16 s 机器人在收腿落脚后持续后仰、向后移动，骨盆逐渐下降，末帧身体倒向红色垫面，没有恢复直立。log step 249 的同文件确定性回放返回 failed=fell，最终评分也记录 fell=1。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：机器人执行有效踢击后，收腿落脚与全身稳定控制未能协调，后仰及后退未被抑制而持续发展成跌倒。log step 215 的生成脚本使用足部目标、腿部逆解、关节PD及质心平衡项；log step 247 更新最终收腿和后方落脚目标并生成交付文件，回放仍为kick后fell，视频 1.44–2.16 s 也显示收腿、落脚期间持续失衡。该证据支持控制层失败，但不能单独确定是逆解/轨迹跟踪、平衡反馈还是某个增益参数所致。",
+    "source_zh": "日志 / 视频 / 评分器字段"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志；25 fps场边视频不能独立测量毫秒级接触时长、精确接触点和冲量，这些数值采用最终评分字段，并由最终交付回放结果佐证。缺少交付轨迹的关节目标与实测值、质心速度及力矩逐帧记录，无法核验agent在log step 256中关于收腿滞后和制动能力的定量解释，故根因仅到控制层。"
+ },
+ "summary_zh": "机器人右脚踢中蓝色护具，评分冲量9.15 N·s、接触时长0.044 s均通过踢击要求，但收腿后持续后仰、后退，至视频 2.16 s 倒向垫面，未完成踢后站稳，完整节点终值为0/1。断点为“踢中护具并站稳”的操作中，现象是全身平衡丧失，证据支持收腿落脚与稳定控制未协调。没有逐帧状态日志，精确接触指标依评分器，控制子类及agent所述定量失衡机制尚不能核实。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "节点未达成（kick阶段通过、站稳未成立）、断点、平衡现象和控制根因均有视频与评分字段支持；未发现需修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:44:51Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--kk2pkmp/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--kk2pkmp/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "3d2b29a0169eef321538fd65bb821d4d0f38cb2a544c2fac5ea6e343afbf1e5c",
+   "video_sha256": "7e1255c5d5ef62956ec4467342b4cc3527207649ec1f1d672b6928dbca0df983",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--tanjnvr.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--tanjnvr.json
new file mode 100644
index 0000000000000000000000000000000000000000..f26a0258f88c4498dee7756d5f0eabf2b7ff3f6e
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--tanjnvr.json
@@ -0,0 +1,201 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-privileged--tanjnvr",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 0,
+  "reward": 0.5,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 1500,
+   "sim_steps": 483,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 0.5,
+   "stages_reached": 1,
+   "stages_total": 2,
+   "fell": 1,
+   "replay_s": 4.09,
+   "bad_trajectory": 0,
+   "metric_contact_s": 0.014,
+   "metric_impulse_n_s": 12.76,
+   "metric_peak_force_n": 1183.2,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 57,
+   "agent_resets": 76,
+   "agent_replays": 1,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.743,
+   "peak_processes": 11,
+   "peak_disk_gb": 0.242,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--tanjnvr/replay.mp4",
+  "duration_s": 4.84,
+  "fps": 25.0,
+  "frames": 121,
+  "sha256": "185632de166877e1c273da1ee6cf6d443b51c53840b7dc48bce5babff014af7b"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 4.04,
+      "log_step": 144,
+      "text_zh": "观察：视频 3.68–4.16 s 中机器人以左脚支撑红色垫面，抬起并伸出右脚击向假人蓝色护具，视频 4.04 s 右脚已到护具前表面。log step 144 对已保存轨迹的新进程重放报告 stages=[kick]、冲量12.76 N·s、接触0.014 s；inputs/run.json 的评分字段给出相同数值，踢击阶段通过。"
+     },
+     {
+      "t": 4.8,
+      "log_step": 153,
+      "text_zh": "观察：视频 4.20–4.80 s 中机器人收回右腿后未建立稳定双脚站姿，身体向后移动并持续下沉，视频 4.80 s 骨盆已接近垫面、躯干明显倾斜。评分器报告 fell=1、stages_reached=1/2、success=0；log step 153 最后重新保存的轨迹同样在仿真 4.83 s 因骨盆高度0.339 m判跌倒。"
+     },
+     {
+      "t": 4.8,
+      "log_step": null,
+      "text_zh": "推断：完整节点要求有效踢击之后连续双脚站稳2.0秒；本次视频踢后立即进入失衡且没有恢复，故此节点从未达到。成功参考视频 1.80–4.04 s 可见机器人收腿后双脚持续落地且躯干竖直，本次末段不具备该可见状态；参考仅用于辨认完成形态。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 4.04 s 附近机器人右脚踢到假人的蓝色胸腹护具，评分器的冲量12.76 N·s及kick阶段通过支持实际触靶。候选名称“击打”；接触方式表没有踢击对应项，按S16记其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 3.68–4.40 s 机器人抬右腿踢击和收腿时，左脚承重支撑红色垫面；此后身体后倾下沉，未形成持续稳定的双脚支撑。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "value": 12.76,
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better"
+  },
+  {
+   "name": "contact_s",
+   "value": 0.014,
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "protector_kicked_and_stood",
+   "t": 4.4,
+   "log_step": null,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 4.04 s 附近机器人右脚击中护具，视频 4.20–4.40 s 收腿期间身体逐渐后退、倾斜，视频 4.60–4.80 s 骨盆迅速下降且未恢复站立；评分器以fell=1结束本次轨迹。失败发生在完整踢击动作的落地恢复过程中。"
+  },
+  "unmet": {
+   "protector_kicked_and_stood": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 4.4,
+    "log_step": null,
+    "evidence_zh": "观察：视频 4.20–4.80 s 机器人踢后收腿，躯干向后倾，骨盆持续下降，末帧已接近地面；inputs/run.json 评分器字段fell=1。机器人未完成落地站稳，且本段失衡没有修复。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：log step 141 的交付生成脚本采用lean→shift→chamber→strike→retract→land→stand关键帧及关节/质心控制，目标包含收腿、落地和站稳；log step 144 保存文件重放及log step 153 最终重新保存均通过kick后跌倒，与视频 4.20–4.80 s 右腿回收时身体后退、下沉一致。该运动控制未能在踢击后及时恢复稳定支撑，导致倾倒；现有证据不足以区分轨迹生成与反馈控制的具体贡献。",
+    "source_zh": "日志 / 视频 / 评分器字段"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志，无法独立核算毫秒级触靶时长、精确接触点及质心速度，冲量和时长采用本次评分器字段。日志中的其他测试不作为本次节点进度证据；踢后失稳可归于控制，但不足以确定具体轨迹或反馈项。"
+ },
+ "summary_zh": "本次机器人以左脚支撑、右脚有效击中护具，评分冲量12.76 N·s、接触0.014 s，原生kick阶段通过。视频 4.20–4.80 s 收腿落地期间机器人后退倾倒，踢后恢复控制未能稳定身体，因此“踢中护具并站稳”完整节点未达到，最终0/1。无逐帧状态日志，数值依据评分器，控制失稳的具体子因无法确定。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "标签与视频和评分器字段一致：右脚在视频 4.04 s 击中护具，之后收腿失衡，4.80 s 倒地，节点未达到，断点、平衡现象和控制根因均有证据支持。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:46:31Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--tanjnvr/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--tanjnvr/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "ee374bb78cb8c791bfffd6a7a2b507d96d9d2a2052bc0bfc87f0b589e617e91f",
+   "video_sha256": "185632de166877e1c273da1ee6cf6d443b51c53840b7dc48bce5babff014af7b",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--zppqhin.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--zppqhin.json
new file mode 100644
index 0000000000000000000000000000000000000000..b965c7ac939d63dbbfc67aa7a169add6c36d5c2e
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-privileged--zppqhin.json
@@ -0,0 +1,177 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-privileged--zppqhin",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 284,
+   "sim_steps": 284,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 2,
+   "stages_total": 2,
+   "fell": 0,
+   "replay_s": 6.26,
+   "bad_trajectory": 0,
+   "metric_contact_s": 0.023,
+   "metric_impulse_n_s": 14.87,
+   "metric_peak_force_n": 1195.2,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 1384,
+   "agent_resets": 143,
+   "agent_replays": 40,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 6.616,
+   "peak_processes": 11,
+   "peak_disk_gb": 0.843,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--zppqhin/replay.mp4",
+  "duration_s": 2.88,
+  "fps": 25.0,
+  "frames": 72,
+  "sha256": "a80ad8d5f604d09b11197b34f867397cda2d0546b3044ec3de1569e1310ab191"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [
+     [
+      2.84,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频0.00–0.92 s，机器人抬起右腿，以左脚支撑红色垫面；视频0.36–0.44 s右脚伸向假人蓝色护具下缘，随后视频0.48–0.64 s收腿并落地，双臂在空中摆动，没有用手操作假人。"
+     },
+     {
+      "t": 2.84,
+      "log_step": 388,
+      "text_zh": "观察：视频0.84–2.84 s机器人双脚落在红色垫面、躯干保持直立，与护具分离；已逐帧检查包括视频最后一秒。log step 388中对最终284行交付文件的两次新进程重放均返回success=true、stages=[kick,stand]、相同终态哈希，并记录踢击冲量14.87 N·s、接触时长0.023 s；inputs/run.json评分器同样报告success=1、deterministic=1、stages_reached=2。"
+     },
+     {
+      "t": 2.84,
+      "log_step": 388,
+      "text_zh": "推断：以log step 388所核验的交付轨迹284个10 ms控制步映射到视频2.84 s（视频72帧、25 fps，末帧为视频2.84 s）；其中踢击阶段的仿真0.455 s与视频0.40–0.48 s的触靶收腿相符，最终站稳事件的仿真2.84 s对应视频末帧。故完整节点在视频2.84 s才达成；毫秒级合法接触及连续200控制步站稳依据评分器验收，不能仅由25 fps画面独立量出。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频0.36–0.44 s机器人右脚向蓝色躯干护具短促踢出，随后收腿；评分器报告有效踢击。候选名称“击打”：现有接触方式没有踢击对应值，按S16使用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.32–0.48 s机器人踢出右腿时左脚支撑红色垫面；随后右脚落地，视频0.84–2.84 s双脚共同支撑身体，完成踢后的站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "value": 14.87,
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better"
+  },
+  {
+   "name": "contact_s",
+   "value": 0.023,
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人以右脚短促踢中蓝色护具，收腿落地后站稳，在视频2.84 s完成“踢中护具并站稳”，最终1/1节点达成。评分器记录冲量14.87 N·s、接触时长0.023 s，最终交付文件两次重放均完成kick与stand且终态一致。无逐帧状态日志，合法接触点、毫秒级接触时长及精确站稳计数由评分器验收支持，视频支持动作顺序及末段站立。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "节点在视频2.84 s达成、contact_actual、时钟换算与评分器数值均与视频及run.json一致，未发现需要修改的内容。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:44:18Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--zppqhin/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-privileged--zppqhin/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "7f25fa949aff92aee94b3a1a0429bd1af15e20429b83e41cab0de6ae86dd7e3f",
+   "video_sha256": "a80ad8d5f604d09b11197b34f867397cda2d0546b3044ec3de1569e1310ab191",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--bupo9hl.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--bupo9hl.json
new file mode 100644
index 0000000000000000000000000000000000000000..2c09c20c3f8e8a5a88ff4bca28e02206a0ccbd2f
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--bupo9hl.json
@@ -0,0 +1,199 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-standard--bupo9hl",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 112,
+   "sim_steps": 4711,
+   "calls": 198,
+   "rows": 4711,
+   "resets": 50,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 4973.8,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.671,
+   "peak_processes": 8,
+   "peak_disk_gb": 0.374,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--bupo9hl/replay.mp4",
+  "duration_s": 1.16,
+  "fps": 25.0,
+  "frames": 29,
+  "sha256": "8329257703e4646c77acdb3afe60a3ba6a0fc869189ab0f0cbd5a664916bb1f8"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": null,
+      "text_zh": "观察：逐帧查看视频0.00–1.12 s；机器人先以左脚支撑红色垫面，视频0.32–0.44 s右腿抬起伸向假人，右脚在视频约0.40–0.44 s到达蓝色护具下部，随后收腿。视频0.56–0.64 s开始明显后仰，视频0.84–1.12 s骨盆持续降低并向后倒下，结尾未恢复直立，更未完成踢后连续双脚站稳。"
+     },
+     {
+      "t": 1.12,
+      "log_step": 702,
+      "text_zh": "观察：log step 702先输出现场服务的Reset to episode: 51，最终outcome: fell、stages: 1 of 2、control_steps: 112；该服务的stages是裁判验收阶段，说明踢击通过、站稳未通过。log step 703再次读取现场服务，episode=51、control_steps=112、done=True，和reward中的n_actions=112、graded_by_live=1一致。"
+     },
+     {
+      "t": 1.12,
+      "log_step": 702,
+      "text_zh": "推断：log step 27的服务说明规定只评分最后回合；log step 702导入带有顶层reset和submit的控制脚本，实际启动并结束第51回合，因此其现场输出对应本次视频。log step 17给出的控制步长为0.01 s，112步对应视频1.12 s，已用视频1.12 s机器人向后倒下的末帧核对；同一log step后半带[ep47_kick]前缀的数值来自agent本地仿真，不用于本回合接触和时刻判定。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频约0.40–0.44 s机器人右脚短促击中假人的蓝色躯干护具后收回；候选名称“击打”，现有接触方式没有踢击对应项，故用其他。现场裁判在log step 702记踢击阶段通过。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.24–0.48 s右腿抬起踢靶时左脚接触红色垫面承重；视频约0.60 s右脚回落，随后双脚附近仍接触垫面，但机器人持续后仰，未建立稳定站姿。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "value": null,
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better"
+  },
+  {
+   "name": "contact_s",
+   "value": null,
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "protector_kicked_and_stood",
+   "t": 0.6,
+   "log_step": 702,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约0.60 s机器人踢后收腿落地时明显向后倾斜，此后至视频1.12 s持续后倒，始终未恢复站稳；log step 702现场裁判以fell结束回合，且仅通过1/2个原生验收阶段。断点是正在执行的完整踢击并站稳步骤。"
+  },
+  "unmet": {
+   "protector_kicked_and_stood": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 0.6,
+    "log_step": 702,
+    "evidence_zh": "观察：视频0.60–0.76 s机器人收腿后躯干明显后仰、双臂前伸，视频0.84–1.12 s后倾加重并向后倒下，骨盆降至贴近垫面；log step 702现场服务最终报告outcome=fell，未出现恢复站姿。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "反馈控制",
+    "evidence_zh": "推断：log step 659中的控制脚本在踢后读取骨盆IMU姿态，以髋踝目标修正执行闭环平衡；log step 702导入并实际执行该脚本，现场输出显示control step 62、82、102的pitch依次为−16.5°、−23.8°、−42.7°，踝修正持续达到−0.250 rad限幅、髋修正随后达到−0.220 rad限幅，仍未抑制后倾并最终跌倒。以0.01 s控制步长对应视频约0.62、0.82、1.02 s，其后倾趋势与视频相符，支持踢后平衡反馈未能稳定本体；不足以进一步断定某一增益、符号或动力学参数错误。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志；run.json的reward未报告本次现场回合的冲量与接触时长，25 fps视频也无法精确量出毫秒级触靶时长，两个原生指标均为null。log step 702后半的本地仿真报告零冲量和更早跌倒，与现场裁判及视频不一致，不作为评分回合证据；agent在log step 713自述的冲量约11–13 N·s也不替代评分器测量。控制记录支持踢后反馈未稳住本体，具体失稳动力学和参数责任仍不能确定。"
+ },
+ "summary_zh": "最后评分回合中，机器人以左脚支撑、右脚踢中护具，现场裁判确认踢击阶段通过，但完整“踢中护具并站稳”节点未达到，最终为0/1。视频约0.60 s收腿落地后机器人持续后仰，至视频1.12 s向后倒下；现场IMU反馈与限幅修正记录支持踢后平衡反馈未能稳定本体。无逐帧状态日志，本回合冲量和触靶时长未由评分器报告，本地仿真及agent自述数值不用于补齐。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频显示右脚踢中护具后收腿，约0.6 s起持续后仰至1.12 s倒地；节点未达成、断点、平衡现象和反馈控制根因与证据一致。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:57:09Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--bupo9hl/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--bupo9hl/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "a7f90d3d7c8f0f5a1b74bc103e33a280d92c577054b26f3955cc306d7113b536",
+   "video_sha256": "8329257703e4646c77acdb3afe60a3ba6a0fc869189ab0f0cbd5a664916bb1f8",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--new7adb.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--new7adb.json
new file mode 100644
index 0000000000000000000000000000000000000000..9c125d8655e11b22ca55aa93d31c6176bb01d564
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--new7adb.json
@@ -0,0 +1,201 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-standard--new7adb",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 348,
+   "sim_steps": 1857,
+   "calls": 176,
+   "rows": 1857,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 1741.7,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 0.304,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.883,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--new7adb/replay.mp4",
+  "duration_s": 3.52,
+  "fps": 25.0,
+  "frames": 88,
+  "sha256": "49478aebb2e4e161c98b33d97efbc11a0a18f5e99fc15bc10f4b60fee680a841"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.48,
+      "log_step": 77,
+      "text_zh": "观察：视频 1.44–2.44 s 机器人以左脚支撑并抬起右腿，视频 2.48 s 右足前端抵达假人蓝色护具下部，视频 2.52–2.76 s 收腿、向后落脚。log step 76 明确重置后执行最后一次尝试；log step 104 的机器人服务裁判回执为 episode=6、stages_reached=1、stages_total=2、outcome=fell，确认有效踢击阶段已通过，完整动作尚未通过。"
+     },
+     {
+      "t": 3.0,
+      "log_step": 104,
+      "text_zh": "观察：视频 2.80–2.88 s 右脚已落回红色垫面，但视频 2.92–3.28 s 机器人继续后移、后仰并抬脚补步，视频 3.32–3.48 s 身体侧转倒下，最后躯干贴近并落到垫面；没有踢后连续双脚站稳的完成区间。log step 104 的服务回执报告最终跌倒且仅通过一个原生阶段。"
+     },
+     {
+      "t": 3.48,
+      "log_step": 104,
+      "text_zh": "推断：最后一次 episode=6 的控制周期为 0.01 s（log step 7 的服务规格及 log step 76、82 的提交参数），log step 104 的 control_steps=348 对应视频 3.48 s，已在末帧倒地事件核对；视频在 25 fps 下共 88 帧、容器时长 3.52 s。此映射只用于该现场尝试的控制步，不使用 agent 工作时间。DAG 将踢击与连续站稳合为一个事件节点，原生阶段通过 1/2 不构成该节点完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 2.48 s 机器人右足踢到假人蓝色躯干护具下部，随后迅速收回；log step 104 的裁判回执确认踢击阶段通过。候选名称“击打”；现有接触方式没有踢击对应值，依 S16 使用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 1.60–2.60 s 右腿抬起和踢靶时左脚支撑红色垫面，视频 2.80–2.88 s 右脚随后落地形成双脚支撑；后续补步仍未稳住身体，视频 3.48 s 躯干也落到垫面。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "value": null,
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better"
+  },
+  {
+   "name": "contact_s",
+   "value": null,
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better"
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "protector_kicked_and_stood",
+   "t": 2.96,
+   "log_step": 85,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 2.92–3.08 s 机器人踢后落地却持续向后倾斜，视频 3.12–3.28 s 抬脚补步也未恢复，视频 3.32–3.48 s 倒向垫面。log step 104 的机器人服务裁判最终报告 outcome=fell、stages_reached=1/2；机器人仍在完成踢后站稳这一完整动作，尚未达成连续站稳验收。"
+  },
+  "unmet": {
+   "protector_kicked_and_stood": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 2.96,
+    "log_step": 104,
+    "evidence_zh": "观察：视频 2.92–3.28 s 机器人后仰越来越明显并补步，视频 3.32–3.48 s 身体侧转倒下，末帧躯干落到红色垫面；log step 104 的服务裁判回执以 fell 终止最后一次 episode=6。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "反馈控制",
+    "evidence_zh": "推断：机器人完成有效踢击后，恢复控制未能抑制身体后移和倾倒。log step 78、80 记录 agent 的控制脚本继续调整右脚落点，log step 82、85 切换 walk_agile 恢复控制，log step 88、91、95 再根据观测调整补步，log step 97、99、102 又提交紧急关节姿态；对应视频 2.80–3.48 s 的身体仍持续后仰并倒下。目标是踢后落脚站稳，实际执行已落脚且尝试纠偏，却未稳定身体，因此归为反馈控制；现有证据不足以指定某一增益或某一次切换为唯一原因。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "观察：inputs/ 中有本次完整视频和运行日志，但没有逐帧状态日志；run.json 的 reward 及可见服务回执未报告 impulse_n_s、contact_s 数值，故两项原生指标留空。25 fps 视频无法精确测定毫秒级足靶接触时长及法向冲量，踢击合法性由最后一次服务裁判的阶段回执支持；也无法进一步隔离导致恢复失稳的具体控制参数或模型误差。"
+ },
+ "summary_zh": "观察：最后一次评分尝试中，机器人以左脚支撑、右脚踢中护具后收腿落地；log step 104 的服务裁判确认原生阶段通过 1/2，但“踢中护具并站稳”完整节点未完成，最终为 0/1。观察：机器人在视频约 2.92 s 起持续后仰，补步后仍于视频 3.48 s 倒地；推断：踢后恢复控制未能稳定身体，根因归为反馈控制。观察：没有逐帧状态日志或本次冲量、接触时长数值，无法进一步精确核验接触指标或定位具体失稳参数。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标签与视频和 log 一致：踢击阶段通过（服务回执 stages_reached=1/2），随后后仰并倒地（outcome=fell），节点未完成；断点、平衡现象与反馈控制根因有证据支持，控制步到视频时间的映射已在末帧核对。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:51:38Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--new7adb/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--new7adb/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "5bfa7911ec0b2e890129500ed3ad6def2262b871e49ee793364aa6dbb6eb0993",
+   "video_sha256": "49478aebb2e4e161c98b33d97efbc11a0a18f5e99fc15bc10f4b60fee680a841",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--o3baf4x.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--o3baf4x.json
new file mode 100644
index 0000000000000000000000000000000000000000..0effb9285650e80c5208cdbf695327310a9e096c
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--o3baf4x.json
@@ -0,0 +1,196 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-standard--o3baf4x",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 178,
+   "sim_steps": 838,
+   "calls": 61,
+   "rows": 838,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2919.0,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 6.678,
+   "peak_processes": 16,
+   "peak_disk_gb": 1.356,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--o3baf4x/replay.mp4",
+  "duration_s": 1.8,
+  "fps": 25.0,
+  "frames": 45,
+  "sha256": "c89182f94d3495688fea0043cdc47d04290e6f1b7322b5bc0f2b007f56648056"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.08,
+      "log_step": 304,
+      "text_zh": "观察：视频 0.64–1.12 s，机器人以左脚支撑红色垫面，抬右腿并伸脚触及假人蓝色护具的下部，随后屈膝收脚。log step 304 的现场服务裁判反馈为“done: fell failed 1 / 2”，即接受踢击阶段但未完成站稳阶段；该步末尾的服务状态为 episode 6、control_steps 178，与本视频的单次踢击后倒下过程一致。"
+     },
+     {
+      "t": 1.4,
+      "log_step": 304,
+      "text_zh": "观察：视频 1.24–1.44 s，机器人右腿下放时躯干持续向远离假人的方向后仰；视频 1.48–1.76 s，身体继续降低直至近乎横躺在垫面上，没有出现踢后连续双脚站稳的窗口。完整“踢中护具并站稳”节点始终未达到。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频约 1.04–1.12 s，机器人右脚伸向并击中假人蓝色躯干护具下部，随后收回；log step 304 的现场裁判接受 kick 阶段。候选名称“击打”：接触方式表无踢击专用值，按 S16 使用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.64–1.28 s，机器人右腿抬起踢击时，左脚接触红色垫面承重；随后右脚下放，视频约 1.40–1.56 s 双脚接近或接触垫面，但身体持续后仰，未形成稳定站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "protector_kicked_and_stood",
+   "t": 1.36,
+   "log_step": 304,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 1.36 s，机器人在踢后收腿落地过程中明显后仰，随后至视频 1.76 s 倒下，始终未恢复站稳。log step 304 的现场服务裁判终止原因为 fell，stages_reached 为 1/2，最终状态为 episode 6；故断点是完整踢击动作的恢复阶段。"
+  },
+  "unmet": {
+   "protector_kicked_and_stood": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 1.36,
+    "log_step": 304,
+    "evidence_zh": "观察：视频 1.28–1.44 s，机器人收回右腿时上身向后倾斜并继续远离假人；视频 1.48–1.76 s，躯干逐渐接近水平、身体落向垫面，未能恢复直立。log step 304 的现场服务裁判报告 fell failed 1/2。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：log step 244 的驱动脚本逐段提交预先计算的关节目标，循环仅检查回合是否结束，没有根据躯干姿态或速度修正平衡；log step 304 生成并现场执行最后的关键帧踢击与落地轨迹，服务反馈踢击通过但随后摔倒。结合视频 1.28–1.76 s 持续后仰而未恢复支撑的过程，归因于踢后恢复控制未能稳定全身；现有证据不足以在轨迹、接触布局或反馈控制中唯一细分。",
+    "source_zh": "日志"
+   }
+  ],
+  "gap_zh": "无逐帧状态日志或现场接触力、质心记录；本次 reward 未报告 impulse_n_s 与 contact_s，25 fps 视频也不能精确量出毫秒级接触时长与冲量。log step 304 开头的 imp=17.1 等数值来自 agent 的本地模拟测试，不能当作现场评分指标；agent 在 log step 313 将失稳归因于反冲及支撑足尺寸的具体解释仍待核实，因此根因仅标到控制。"
+ },
+ "summary_zh": "本次评分对应 log step 304 的最后一次现场尝试（episode 6）：机器人右脚击中护具，裁判接受踢击阶段（1/2），但完整“踢中护具并站稳”节点未达成，最终为 0/1。视频约 1.36 s，机器人在收腿落地时明显后仰，至视频 1.76 s 倒下，显示踢后恢复控制未能维持平衡。缺少现场状态和力记录，不能精确报告冲量、接触时长或进一步确定失稳机制。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标签与视频一致：视频约1.0–1.16 s右脚触及护具，其后约1.3 s起后仰，至1.76 s倒地；节点未达成、断点在踢后收腿落地阶段（操作中）、现象平衡、根因控制（未细分）均有证据支撑，未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:50:03Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--o3baf4x/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--o3baf4x/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "21415f595a2c5171e88cf9b99bf25092d026e52f47804a3ecb148efbf8e84b45",
+   "video_sha256": "c89182f94d3495688fea0043cdc47d04290e6f1b7322b5bc0f2b007f56648056",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--oatjwqx.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--oatjwqx.json
new file mode 100644
index 0000000000000000000000000000000000000000..9ecca199140cf17badfc6531ce5abb86769db645
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--oatjwqx.json
@@ -0,0 +1,223 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-standard--oatjwqx",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 72,
+   "sim_steps": 144,
+   "calls": 89,
+   "rows": 144,
+   "resets": 1,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 7044.0,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 4.36,
+   "peak_processes": 12,
+   "peak_disk_gb": 2.507,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--oatjwqx/replay.mp4",
+  "duration_s": 0.76,
+  "fps": 25.0,
+  "frames": 19,
+  "sha256": "95d0437b36ae375a69f1dbfcbeea23ff6f87d19f633837fcd4791d9338c912a7"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.56 s，机器人面向假人，双脚在红色垫面支撑，随后下蹲、伸腿起跳；尚未发生踢中护具并落地站稳的完整动作。"
+     },
+     {
+      "t": 0.72,
+      "log_step": 155,
+      "text_zh": "观察：已检查本次视频全部19帧；视频 0.60–0.72 s机器人腾空抬起右腿，末帧右脚抵到黑色底座上缘，低于蓝色白圆标护具，未见合法击靶后的收腿落地与站稳。log step 153重置后，log step 155的现场服务返回episode=2、control_steps=72、sim_time_s=0.72、done=True、outcome为“the kick landed outside the trunk protector”、stages_reached=0/2；此为最后一次现场评分尝试。"
+     },
+     {
+      "t": 0.72,
+      "log_step": 155,
+      "text_zh": "推断：log step 3给出的控制周期为0.01 s，log step 155的本局72个控制步对应仿真0.72 s，与视频0.72 s末帧的抬脚触底座核对一致；节点判为未达来自可见动作与现场裁判的具体判负，非仅据总分为零。成功参考视频0.80–1.40 s可见脚升向护具并收回，之后至参考视频4.04 s持续站立，本次没有出现该完整过程。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频0.00–0.56 s，机器人双脚在红色垫面承重，下蹲后蹬地起跳；这是后续腾空摆腿之前的地面支撑。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频0.72 s，机器人抬起的右脚碰到假人黑色底座上缘，未击中上方蓝色护具；候选名称“击打”，现有接触方式没有踢击对应值。该接触发生在起跳支撑之后。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "protector_kicked_and_stood",
+   "t": 0.72,
+   "log_step": 155,
+   "where": "操作中",
+   "evidence_zh": "观察：视频0.64–0.72 s，机器人正在腾空抬右腿踢靶，右脚在末帧抵达黑色底座上缘，低于护具；log step 155的现场裁判以“the kick landed outside the trunk protector”终止本局，kick与stand两个原生阶段均未通过。"
+  },
+  "unmet": {
+   "protector_kicked_and_stood": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 0.72,
+    "log_step": 155,
+    "evidence_zh": "观察：视频0.64–0.72 s机器人腾空抬右腿，视频0.72 s右脚触到假人黑色底座上缘，踢击接触位置低于上方蓝色白圆标护具；log step 155的现场服务返回outcome为“the kick landed outside the trunk protector”。机器人尚在空中，评分片段未出现踢后摔倒。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "推断：log step 150的控制脚本以预设关节轨迹和机身IMU反馈执行下蹲、起跳、高抬腿及伸膝击靶，log step 153修正读取反馈并改为每控制步提交；实际视频0.64–0.72 s的抬脚路径仍先碰底座，未实现脚部越过底座后击中护具，支持执行控制失败。log step 155还记录waist_roll_q越界被限幅，但没有足够证据确定限幅、反馈律或轨迹参数各自的因果贡献，故不细分控制子类。"
+   }
+  ],
+  "gap_zh": "观察：inputs/没有状态日志；25 fps单一场边视角不足以核验毫米级接触点、冲量和接触时长，评分字段也未报告impulse_n_s或contact_s，因此两项原生指标留空。log step 158所写的踢后摔倒及有效冲量来自agent自建仿真，不是本次现场评分的事实；缺少现场逐步跟踪误差，不能进一步确定控制根因子类。"
+ },
+ "summary_zh": "观察：评分视频中机器人完成下蹲起跳和抬右腿，但视频0.72 s右脚碰到黑色底座上缘，现场裁判在log step 155判为踢中护具外；“踢中护具并站稳”最终0/1。推断：摆腿与机身控制未让脚部越过底座到达护具，断点在该节点操作中，现象为位姿不准，根因归控制。观察：没有状态日志或本局冲量、接触时长读数，具体控制失配无法细分；agent自建仿真的踢后摔倒不计入本次评分轨迹。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 6,
+    "summary_zh": "时钟、末局（episode 2）、节点未达、断点基本正确；但现象“被挡/碰撞”不符合定义：脚没有被阻挡，而是瞄准的踢击落点偏低、先碰到底座，属位姿不准。",
+    "n_issues": 1,
+    "n_major": 1
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "时钟、末局（episode 2、72控制步=0.72 s）、节点未达、断点（操作中）、位姿不准现象和控制根因均与视频及log step 155一致；第1轮意见已采纳。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:03:19Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--oatjwqx/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--oatjwqx/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--oatjwqx/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--oatjwqx/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "88b383ad526aefa3f19d72e7284774b1fb96709345bce5315ffb980e22e849de",
+   "video_sha256": "95d0437b36ae375a69f1dbfcbeea23ff6f87d19f633837fcd4791d9338c912a7",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--qrd98rx.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--qrd98rx.json
new file mode 100644
index 0000000000000000000000000000000000000000..4f5f9f450fa59493fbf62d6b4a651e682a91f07b
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--qrd98rx.json
@@ -0,0 +1,218 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-standard--qrd98rx",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 163,
+   "sim_steps": 163,
+   "calls": 6,
+   "rows": 163,
+   "resets": 0,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 4072.7,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.408,
+   "peak_processes": 11,
+   "peak_disk_gb": 0.447,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--qrd98rx/replay.mp4",
+  "duration_s": 1.64,
+  "fps": 25.0,
+  "frames": 41,
+  "sha256": "06bd11c1e098dcedc54871072a88c9d586c9557814a6a5bfc3155e34f86664d8"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.2,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 0.00–1.60 s，机器人先站在假人前，约视频 1.08 s 开始明显抬右腿，视频 1.20–1.56 s 屈膝抬脚，左脚支撑红色垫面。"
+     },
+     {
+      "t": 1.6,
+      "log_step": 194,
+      "text_zh": "观察：视频 1.40–1.60 s 机器人右腿屈膝后向假人伸出，脚部轮廓从黑色底座上方朝蓝色护具下缘靠近，右脚尚未收回落地；末帧不足以直接确认脚与护具的接触点。log step 194 的现场机器人服务返回 outcome=\"the protector was pushed, not struck\"、stages_reached=0/2、done=true。推断：依据 log step 189 的控制周期0.01 s及 log step 194 的163个控制步，终止对应视频约 1.63 s，与视频 1.60 s 抬腿伸向假人后立即结束相符；服务判定支持护具接触，但未完成有效踢击后双脚站稳的完整节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 1.40–1.60 s 机器人抬起右脚向假人伸出，视频 1.60 s 脚前端接近蓝色护具下缘，具体接触点帧上不可确认；log step 194 现场服务返回“the protector was pushed, not struck”。推断：结合右腿动作及服务判定，本局发生右脚对护具的接触，但被判为推压。候选名称“击打”，接触方式表无踢击对应值，按S16使用其他，不表示有效踢击已经成立。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 1.20–1.60 s，机器人左脚承重支撑红色垫面，右脚抬起向护具运动。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "protector_kicked_and_stood",
+   "t": 1.6,
+   "log_step": 194,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 1.60 s 机器人正抬右腿向假人伸出，仍为左脚单脚支撑，未进入收腿落地后的站稳阶段；log step 194 现场服务在控制步163以“the protector was pushed, not struck”判负并结束本局，stages_reached=0/2。脚与护具的具体接触点未由末帧直接确认。"
+  },
+  "phenomena": [
+   {
+    "cat": "接触",
+    "item": "接触力异常",
+    "t": 1.6,
+    "log_step": 194,
+    "evidence_zh": "观察：log step 194 的现场机器人服务明确返回“the protector was pushed, not struck”，将本局接触判为推压，未接受为短促击打；这是服务对唯一现场局的判定，按N5作为观察使用。视频 1.40–1.60 s 仅见机器人右脚向假人伸出，前端接近蓝色护具下缘，不能从这些帧确认护具上的接触点或接触时长。推断：以服务判定为依据保留“接触力异常”，不以画面证明推压持续时间。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：log step 190 的控制脚本生成关节目标关键帧及收腿段，log step 194 确认这些指令实际执行163个控制步，并由现场服务判为推压；结合视频 1.40–1.60 s 的右腿伸出动作，本次踢腿控制未实现裁判所需的短促击打。没有现场接触力、目标跟踪或毫秒级轨迹记录，具体接触点也不能由末帧确认，无法可靠细分为落点、轨迹、反馈或具体参数问题。",
+    "source_zh": "日志"
+   }
+  ],
+  "gap_zh": "本局无状态日志，25 fps视频不能精确测量冲量、接触时长及踝局部接触点；评分器reward未报告impulse_n_s或contact_s，故原生指标值留null。log step 260的“六次尝试”及恢复阶段摔倒为agent自述；实际 log step 194 后的提交（如 log step 215、241、247）均返回409 episode finished，始终episode=1、resets_left=5，评分器resets=0，故本次只标首局现场执行，不将后续离线调参结果或离线摔倒移入视频。agent关于需扣除分离确认时长的规则解释也未经现场原始数据验证。 复查视频 1.40–1.60 s 放大帧：右脚部分轮廓低于护具，前端在末帧接近护具下缘，不能直接确认接触点；不采纳审查中“脚明显远低于护具”的确定表述。踢击落点偏低是否参与失败、与服务判定推压的关系，缺少现场接触点和时序记录，不能可靠区分，故不另加“位姿不准”。依据DECISIONS.md N5的Now，唯一现场局的服务判定仍按观察使用，保留其支持的接触力异常。",
+  "unmet": {
+   "protector_kicked_and_stood": false
+  }
+ },
+ "summary_zh": "机器人抬右腿向假人伸出，但现场服务在控制步163（对应视频约 1.63 s）以推压而非击打判负，未完成“踢中护具并站稳”，终态0/1。断点为踢击操作中的接触力异常，依据现场服务判定，推断踢腿控制未实现所需短促击打；本次视频未出现收腿落地或摔倒。画面只能确认脚前端接近护具下缘，不能直接确认接触点及接触时长，落点与推压的关系仍不确定，置信度由8降至6；不采用agent离线数值或后续被拒绝的提交作为本局进度。",
+ "conf": 6,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 6,
+    "summary_zh": "节点未达成、断点、时钟均可接受；但标签把“视频 1.60 s 右脚触到蓝色护具”当作观察，而末几帧（1.40–1.64 s）显示脚尖仍在黑色底座上缘高度、明显低于蓝色护具，未见接触护具；该观察应降为“服务判定的推断”，并改写相关证据与接触描述。",
+    "n_issues": 2,
+    "n_major": 1
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点未达成、断点（踢击操作中）、时钟（控制步163≈视频1.63 s，已声明为推断）均与视频和 log step 194 的服务读数一致；第1轮提出的“脚触护具”观察已改为推断并在 gap_zh 中说明落点不确定，无新证据需改动。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:02:52Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--qrd98rx/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--qrd98rx/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--qrd98rx/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--qrd98rx/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "5a99be8f8a78ac1719bdc5c99ac64b21068a771da51ae7352d6a2635f9b290de",
+   "video_sha256": "06bd11c1e098dcedc54871072a88c9d586c9557814a6a5bfc3155e34f86664d8",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--ruka6kb.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--ruka6kb.json
new file mode 100644
index 0000000000000000000000000000000000000000..0c9ff5215cf47db6fc10188e1b028fecae00316a
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--ruka6kb.json
@@ -0,0 +1,182 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-standard--ruka6kb",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "live_success": 1,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 912,
+   "sim_steps": 2690,
+   "calls": 1737,
+   "rows": 2690,
+   "resets": 4,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 2263.7,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 2.708,
+   "peak_processes": 15,
+   "peak_disk_gb": 0.371,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--ruka6kb/replay.mp4",
+  "duration_s": 9.16,
+  "fps": 25.0,
+  "frames": 229,
+  "sha256": "7698463a65f4cdba50cff792f3ee6d7af0da7134e8f24d42ef033a2c84a7cef4"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [
+     [
+      9.12,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 3.08,
+      "log_step": null,
+      "text_zh": "观察：视频 2.96–3.04 s 机器人右膝抬高，视频 3.08 s 右脚向前伸出并碰到假人蓝色护具，左脚留在红色垫面支撑；视频 3.12–3.40 s 右脚已离开护具并收回。"
+     },
+     {
+      "t": 3.84,
+      "log_step": null,
+      "text_zh": "观察：视频 3.44–3.68 s 机器人放下右腿，视频 3.76–3.84 s 右脚落到红色垫面，随后调整双腿站姿；视频 7.04–9.12 s 双脚着地、躯干保持直立，与假人之间有间隔，未见倚靠假人或其他部位撑地。"
+     },
+     {
+      "t": 9.12,
+      "log_step": 111,
+      "text_zh": "观察：log step 111 的机器人服务输出显示重置后 episode=5，从 control_steps=0 开始，末尾 stand 返回 done=true、reason=success、stages_reached=2，最终 control_steps=912；log step 113 再次读取相同成功状态。评分字段 live_success=1、graded_by_live=1、finalized=1；前四次测试不属于本视频的评分尝试。"
+     },
+     {
+      "t": 9.12,
+      "log_step": 111,
+      "text_zh": "推断：log step 5 服务规格给出 control_dt_s=0.01，log step 111 本次评分的 912 个控制步对应视频约 9.12 s，与视频总长 9.16 s 及已检查的视频 9.12 s 末帧一致；该步脚本的踢击检测索引 ic=308 对应视频约 3.08 s，已在视频 3.08 s 的触靶画面核对。完整踢击并连续站稳节点以裁判成功时刻约视频 9.12 s 记为达成；不能仅在触靶或刚落地时提前记完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 3.08 s 机器人右脚击中假人的蓝色躯干护具，随后立即收回。候选名称“击打”；现有接触方式无踢击对应项，按 S16 记为其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 2.96–3.40 s 踢击期间机器人左脚支撑红色垫面；收腿后右脚在视频约 3.76–3.84 s 落地，末段双脚共同支撑垫面。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": null,
+ "summary_zh": "评分采用第5次现场尝试：机器人在视频约 3.08 s 用右脚短促踢中蓝色护具，收腿落地后调整站姿，完整“踢中护具并站稳”节点在视频约 9.12 s 达成，最终为1/1。log step 111 的机器人服务确认成功、完成2个裁判阶段，视频末段双脚站立且脱离假人，与成功判定一致。没有状态日志，评分字段未报告本次实际冲量和接触时长，故两项原生指标留空；数值门槛及连续站稳计数依据本次裁判成功结果确认。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "评分成功，节点 protector_kicked_and_stood 达成、final=true；视频 3.08 s 右脚击中护具、9.12 s 站稳，接触标签与 S16 一致，无失败层。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:46:19Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--ruka6kb/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--ruka6kb/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "fc062b4c1dadf2625312334ba46e2eb415a20a9153b474658f75c2d5275ed9de",
+   "video_sha256": "7698463a65f4cdba50cff792f3ee6d7af0da7134e8f24d42ef033a2c84a7cef4",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--vyrfjem.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--vyrfjem.json
new file mode 100644
index 0000000000000000000000000000000000000000..a89d17e873f84c50d3dc93e99af05de7fc5ba97c
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--vyrfjem.json
@@ -0,0 +1,172 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-standard--vyrfjem",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 390,
+   "sim_steps": 390,
+   "calls": 12,
+   "rows": 390,
+   "resets": 0,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 3761.2,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.102,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.323,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--vyrfjem/replay.mp4",
+  "duration_s": 3.92,
+  "fps": 25.0,
+  "frames": 98,
+  "sha256": "b982f3b14afc8ff1f539ff5c98db954c279b668a59f8e6280928a42167903373"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.96,
+      "log_step": 39,
+      "text_zh": "观察：本次视频 0.00–3.88 s 的全部98帧中，银色机器人仅站立、屈膝和调整姿态，脚始终远低于蓝色白圆标护具，未见向护具踢击或接触假人；视频 2.92–3.88 s 末段逐帧可见机器人恢复直立，仍与假人分离。log step 25、27、30、39、42 的七次现场提交只有站立、初始关节姿态和零速度步态测试，没有踢击指令。"
+     },
+     {
+      "t": 3.88,
+      "log_step": 138,
+      "text_zh": "观察：log step 138 的现场服务状态为 episode=1、control_steps=390、done=false、reason=null，与评分记录 n_actions=390、resets=0 及本次短视频相符。log step 144 的踢击与站稳结果来自代理自建 LocalSim，并非该现场回合，不能计作本节点进度。"
+     },
+     {
+      "t": 0.96,
+      "log_step": null,
+      "text_zh": "观察：成功参考视频 0.52–1.12 s 清楚显示机器人抬腿、足端伸向蓝色护具再回收，参考视频 3.00–4.04 s 显示踢后的双脚站立；这些可见动作在本次视频中均未出现，仅用于确认该视角能够显示踢击。推断：本次未满足完整踢击并站稳判据，节点全程为假，而非因遮挡无法判断。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "unknown",
+   "t": null,
+   "log_step": 147,
+   "where": "首步前",
+   "evidence_zh": "观察：视频 0.00–3.88 s 仅含姿态测试，没有踢击；log step 138 的现场服务仍为 done=false、reason=null。log step 144 仅验证自建 LocalSim，log step 146 只记录方案笔记，随后 log step 147 返回 Gemini 过滤拦截提示，未见把踢击方案提交现场服务或现场回合自然终结。推断：评分结果固定前发生外部过滤中断，没有可定位的机器人踢击失败事件，按外部中断规则使用 unknown／首步前。"
+  },
+  "unmet": {
+   "protector_kicked_and_stood": false
+  },
+  "phenomena": [],
+  "root_causes": [],
+  "gap_zh": "观察：log step 147 明示 Gemini 过滤拦截；最后现场状态（log step 138）仍为 done=false，之后没有现场动作提交或完成验收记录，评分表 success=0 不表示机器人已经物理判负。推断：这是结果固定前的外部中断，应重测并排除出代理失败归因，因此现象、根因均留空，不标没尝试；自建 LocalSim 中的碰撞、失衡和成功均不属于本次评分视频。输入没有状态日志，且本次评分字段未报告 impulse_n_s 或 contact_s，两项原生指标留 null。"
+ },
+ "summary_zh": "本次评分视频中机器人仅完成站立、屈膝和原地姿态测试，没有踢到护具，“踢中护具并站稳”最终为0/1。代理自建 LocalSim 的踢击结果未提交到现场回合，现场服务尚未结束时出现 Gemini 过滤中断（log step 138、144、147），故保留目录 success=0，但按外部中断处理，不归因于机器人失败。无状态日志和本次原生接触指标。",
+ "conf": 9,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "标注符合外部过滤中断规则：视频无踢击，节点为假；log step 147 为 Gemini 过滤拦截，现场回合未提交踢击方案（踢击仅在自建 LocalSim 中验证），因此 breakpoint unknown/首步前，现象与根因留空并有 gap_zh。未发现需修改之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:48:12Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--vyrfjem/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--vyrfjem/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "cd5b9b55fd7b29e6cd92ed7132dd62cac36ba607e33a78795a6b74bf6c36dc6a",
+   "video_sha256": "b982f3b14afc8ff1f539ff5c98db954c279b668a59f8e6280928a42167903373",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--x78ydzo.json b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--x78ydzo.json
new file mode 100644
index 0000000000000000000000000000000000000000..7643b09ab2d9f06da3602a05b972cc27bcdc087c
--- /dev/null
+++ b/phase2/embodied-olympics-fun/taekwondo_kick/trajectories/embodied-olympics-fun-taekwondo-kick-i00-standard--x78ydzo.json
@@ -0,0 +1,201 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-taekwondo-kick-i00-standard--x78ydzo",
+ "taskKey": "embodied-olympics-fun/taekwondo_kick",
+ "dag_sha256": "63ae944f698e6469950e0e47b5b44e99a40b4ebb86df1180bd7baa564cd727b9",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 404,
+   "sim_steps": 17378,
+   "calls": 1159,
+   "rows": 3099,
+   "resets": 50,
+   "reset_budget": 50,
+   "sim_startups": 1,
+   "elapsed_s": 3644.3,
+   "video_rendered": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--x78ydzo/replay.mp4",
+  "duration_s": 4.08,
+  "fps": 25.0,
+  "frames": 102,
+  "sha256": "378fcd76fdaaf7b8c73bbbfbbfb58783d2fcecf56c1d6d61443afc16f1f073e9"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "protector_kicked_and_stood": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 3.4,
+      "log_step": null,
+      "text_zh": "观察：视频 3.12–3.32 s，机器人以左脚支撑、抬起并展开右腿；视频 3.40–3.56 s，右脚到达假人胸腹前蓝色白圆标护具，随后右腿下收。"
+     },
+     {
+      "t": 3.76,
+      "log_step": null,
+      "text_zh": "观察：视频 3.64–3.80 s，机器人收腿时躯干向后倾斜、骨盆下沉；视频 3.84–4.04 s，后仰持续加剧并倒在红垫上，未形成踢后连续双脚站稳窗口。"
+     },
+     {
+      "t": null,
+      "log_step": 210,
+      "text_zh": "观察：log step 210 中 agent 自述最终一次右脚踢击被记为 stage 1、随后恢复时摔倒，未完成站稳；inputs/run.json 的评分字段为 success=0、live_success=0、graded_by_live=1、finalized=1。该自述仅辅助解释视频，未据此赋予冲量或接触时长数值。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "腿/躯干",
+      "how": "其他",
+      "note_zh": "观察：视频 3.40–3.56 s，机器人右脚伸向并触及假人的蓝色护具，随后收回。候选名称“击打”；接触方式表没有踢击对应值，按规则使用其他。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 3.12–3.64 s，机器人抬右腿踢靶时由左脚接触红色垫面承重；视频 3.72–3.80 s，右脚下落到垫面附近，但躯干持续后仰，未恢复稳定双脚站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "impulse_n_s",
+   "unit": "N s",
+   "threshold": 7.2,
+   "direction": "higher_better",
+   "value": null
+  },
+  {
+   "name": "contact_s",
+   "unit": "s",
+   "threshold": 0.05,
+   "direction": "lower_better",
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "protector_kicked_and_stood",
+   "t": 3.72,
+   "log_step": null,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 3.40–3.56 s 机器人右脚到达蓝色护具后开始收腿，视频约 3.72 s 身体明显向后倒，至视频 4.04 s 倒在红垫上；踢击后的落地站稳属于同一节点，其验收从未完成。"
+  },
+  "unmet": {
+   "protector_kicked_and_stood": false
+  },
+  "phenomena": [
+   {
+    "cat": "全身",
+    "item": "平衡",
+    "t": 3.72,
+    "log_step": null,
+    "evidence_zh": "观察：视频 3.64–3.80 s，机器人右腿下收，躯干由直立转为明显后仰，骨盆下沉；视频 3.84–4.04 s 继续向后倒至躯干贴近并落到垫面，未重新站起。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 3.72,
+    "log_step": 208,
+    "evidence_zh": "推断：log step 208 展示的最终控制脚本先发送右腿关节目标序列，收腿时短暂发送初始姿态目标，再切换到后退行走策略；视频 3.64–4.04 s 显示收腿和恢复过程中身体后仰未被纠正，最终摔倒，支持踢击与落地恢复的全身控制失败。现有记录不足以进一步区分轨迹、反馈或具体参数问题。",
+    "source_zh": "日志"
+   }
+  ],
+  "gap_zh": "观察：inputs/ 没有本次状态日志，评分字段也未报告 impulse_n_s 或 contact_s；25 fps 视频不能精确确定踝局部接触点、毫秒级触靶时长或冲量，原生指标均留 null。log step 207、210 对最终一次 stage 1 的说明是 agent 自述，未附最终裁判的细项输出；log step 194 的裁判结果属于较早的 episode 47，未用作本次节点或指标证据。最终脚本在 log step 208 的中段被截断，故根因仅标到控制，不细分。"
+ },
+ "summary_zh": "机器人在视频约 3.4 s 以右脚触及蓝色护具并收腿，但在视频约 3.72 s 开始明显后倒，至视频 4.04 s 已倒在垫上，“踢中护具并站稳”未完成，最终为 0/1。失败现象为全身平衡丧失，结合最终脚本推断踢击后的恢复控制失败；缺少本次冲量、接触时长及详细状态记录，不能独立核验有效踢击阈值或细分控制根因。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频 3.40–3.56 s 右脚触及蓝色护具，3.64 s 起后仰，4.04 s 倒在垫上；节点未完成、断点、平衡现象与控制根因（已标推断、未细分）均有证据支持，未发现需改动之处。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:56:44Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--x78ydzo/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-taekwondo-kick-i00-standard--x78ydzo/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "9d20641574b12b7da01c23c9ae13a88a840864a66fce8c22cdab5bd85fde3344",
+   "video_sha256": "378fcd76fdaaf7b8c73bbbfbbfb58783d2fcecf56c1d6d61443afc16f1f073e9",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--duiew8n.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--duiew8n.json
new file mode 100644
index 0000000000000000000000000000000000000000..a3792476786a61cfe65d27dbb08acd8d4ce3f564
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--duiew8n.json
@@ -0,0 +1,215 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-privileged--duiew8n",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 0,
+  "reward": 0.5,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 491,
+   "sim_steps": 491,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 0.5,
+   "stages_reached": 2,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 5.89,
+   "bad_trajectory": 0,
+   "metric_release_speed_m_s": 13.99,
+   "metric_target_offset_m": 0.234,
+   "metric_wall_speed_m_s": 12.62,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 571,
+   "agent_resets": 54,
+   "agent_replays": 4,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-privileged--duiew8n/replay.mp4",
+  "duration_s": 4.92,
+  "fps": 25.0,
+  "frames": 123,
+  "sha256": "713649570c5eb98316c704d181c592add46f34086d61a04c623c46803d6d59a0"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–3.96 s，机器人右手带着黄绿色网球准备、转体并挥臂；视频 4.00 s 球已离手向墙面飞行，视频 4.36–4.44 s 球到达圆靶附近并反向返回。初始握球及投出命中尚不满足完整投接节点。"
+     },
+     {
+      "t": 4.8,
+      "log_step": 400,
+      "text_zh": "观察：视频 4.48–4.88 s 连续帧中，反弹球离墙后不断下降，视频 4.80–4.88 s 已在机器人双手下方，末帧接近小腿高度，未见任一手包住球或随手停留。log step 400 对最终交付文件重新生成并回放，服务均报告 stages=[release,target]、steps=491、failed=the ball touched the ground，且终态哈希相同；评分器 stages_reached=2/4，与画面一致，接住及持球阶段未完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.00–3.96 s，右侧灵巧手围握网球并带球转体挥臂，随后张手投出；视频 4.00 s 球与右手分离。此抓握属于投球准备，反弹后未见抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–4.88 s，机器人双腿和双脚支撑身体，配合躯干转动完成投球；末段仍站立，未见摔倒。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": 0.234
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 13.99
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 12.62
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "wall_throw_caught",
+   "t": 4.8,
+   "log_step": 400,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 4.72–4.88 s，返回网球从机器人前方下降到手部下方，双手没有在球的路径上形成接球，完整投墙接稳球步骤未完成。log step 400 的最终交付回放仅到 release、target，并由裁判以球触地结束；该日志为交付文件的复验，与本视频过程一致。"
+  },
+  "phenomena": [],
+  "root_causes": [],
+  "gap_zh": "没有逐帧状态日志，25 fps 画面不能核实毫秒级接触、掌侧几何或握合细节；视频末帧为视频 4.88 s，球触地的终止来自 log step 400 最终交付复验的裁判记录（仿真 4.91 s），不声称末帧已直接看到触地。log step 397 的交付生成器在 release 后将右臂及躯干目标速度置零，不再生成接球段；log step 400 重新运行该生成器并复验同一终态哈希，视频 4.48–4.88 s 也未见朝返回球的接球运动。依 docs/03_trajectory.md Step 3 和 CASES.md FAIL-23，不能把会话中的接球试验当成交付轨迹内的动态时机失败，故现象留空；完整投接节点已从投球启动，也不标没尝试。接球试验的具体失败原因及保留投靶检查点的理由主要来自 log step 400 的 agent 自述，不能据此确定根因；脚本缺少接球段本身也不足以反推控制根因（Step 3.3、CASES.md FAIL-22），故根因留空。lowcmd 尾部两个 0 的闭合语义未核实，不作握合命令依据；会话中下蹲及逆解试验未进入本次交付回放。",
+  "unmet": {
+   "wall_throw_caught": false
+  }
+ },
+ "summary_zh": "机器人完成右手投出并命中圆靶，评分器记录 2/4 个原生阶段，但未接住反弹球，完整“投墙接稳球”节点最终为 0/1。视频 4.72–4.88 s 球下降到双手下方；log step 397 的交付脚本在释放后不再生成接球段，log step 400 的最终复验由裁判以球触地结束。交付轨迹未包含朝返回球的接球运动，不能标动态时机；缺少独立证据确定未交付接球段的根因，现象和根因均留空。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 6,
+    "summary_zh": "节点值、断点和时间都对（final=false，球触地，stages 2/4）。问题在现象：交付脚本（log step 397/400）出手后右臂速度置零、不含任何接球动作，标“动态时机·没跟上”声称有一次瞄准运动没赶上球，而交付轨迹里并没有这个运动（CASES FAIL-23）。根因“控制”的证据也主要是脚本缺失加推断，需要降低表述。",
+    "n_issues": 2,
+    "n_major": 1
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点 wall_throw_caught=false、断点 ref=该节点、grader 2/4 与帧一致；第1轮后现象和根因留空并在 gap_zh 说明（交付脚本无接球段，原因仅来自 agent 自述），符合 FAIL-22/23，无需再改。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:59:28Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--duiew8n/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--duiew8n/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--duiew8n/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--duiew8n/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "9354e8793e5711ca29796b7b7448fd0659ab92204476cfd5cb96e263288a126a",
+   "video_sha256": "713649570c5eb98316c704d181c592add46f34086d61a04c623c46803d6d59a0",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--ffybj5m.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--ffybj5m.json
new file mode 100644
index 0000000000000000000000000000000000000000..11aff9a44abc10c19089216e1b71ba928746d57f
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--ffybj5m.json
@@ -0,0 +1,215 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-privileged--ffybj5m",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 0,
+  "reward": 0.5,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 238,
+   "sim_steps": 238,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 0.5,
+   "stages_reached": 2,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 4.87,
+   "bad_trajectory": 0,
+   "metric_release_speed_m_s": 7.96,
+   "metric_target_offset_m": 0.189,
+   "metric_wall_speed_m_s": 7.0,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 19,
+   "agent_resets": 8,
+   "agent_replays": 4,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 4.486,
+   "peak_processes": 29,
+   "peak_disk_gb": 0.24,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-privileged--ffybj5m/replay.mp4",
+  "duration_s": 2.4,
+  "fps": 25.0,
+  "frames": 60,
+  "sha256": "424b0ba4258a6c3955aabb60e76ff2b2373667e1414e455bfa2ea7d14e410ab8"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.28,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–1.20 s，机器人双脚支撑地面，初始右手持球后向后下方摆臂；视频 1.24–1.32 s，黄绿色球离开右手向墙面飞行，右臂继续上摆。初始持球及投出不等于完成反弹接稳球。"
+     },
+     {
+      "t": 2.08,
+      "log_step": 221,
+      "text_zh": "观察：视频约 2.04–2.08 s，球到达圆靶附近后反向下落；log step 221 对交付轨迹的回放裁判输出 stages=[release,target]、target_offset_m=0.189，支持出手及命中已完成，但未报告 catch 或 hold。"
+     },
+     {
+      "t": 2.36,
+      "log_step": 221,
+      "text_zh": "观察：视频 2.08–2.36 s，反弹球仍在墙与机器人之间，下降至蓝色地面，始终未回到双手处；机器人右手停在头部附近、左手留在身前，没有反弹后持球画面。log step 221 的交付回放裁判明确返回 the ball touched the ground，且评分器 success=0、stages_reached=2/4；完整投墙接稳球节点未达到。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "机器人初始右手手指与掌部包住网球，持球后摆并挥臂投出（视频 0.00–1.28 s）；之后松开，反弹后没有形成新的抓握。左手未接触球。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "机器人双脚在地面支撑完成后摆、投掷和等待反弹，视频 0.00–2.36 s 保持站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": 0.189
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 7.96
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 7.0
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "wall_throw_caught",
+   "t": 2.08,
+   "log_step": 221,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 2.08 s，网球已从靶区反弹且继续向下，机器人仍在原地，双手未迎向球；视频 2.32–2.36 s，球在远离机器人双手的位置降到地面，未发生接球。log step 221 的交付文件回放仅完成 release、target，裁判以球触地结束；故失败步骤为完整投墙接稳球。"
+  },
+  "unmet": {
+   "wall_throw_caught": false
+  },
+  "phenomena": [
+   {
+    "cat": "接触",
+    "item": "没建立接触",
+    "t": 2.08,
+    "log_step": 221,
+    "evidence_zh": "观察：视频 2.08–2.36 s，网球反弹后在机器人前方远处下落，未与任一手形成接球持握；log step 221 的交付回放无 catch、hold 阶段，并以球触地失败。这里缺少的是反弹后的接球接触，初始右手持球已主动投出，不是接住后又失握。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "认知",
+    "item": "推理/规划",
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "推断：投掷计划只优化命中圆靶，没有把反弹球路与后续手部承接联合规划。log step 215 的交付生成脚本在投掷循环中持续施加固定右臂力矩，左臂使用固定目标；右手开掌后不再闭合，左手始终张开，没有接球阶段或反弹后拦截动作。log step 220 只修改肩偏航力矩并生成最终文件，log step 221 验证该文件只到 target；视频 2.08–2.36 s 也显示双手未迎球，球在到手前落地。log step 228 中 agent 自述以固定站位推算接球不可达，因此保留这一仅完成投墙的方案；这是跨阶段策略不足，而不是由其停止本身推定根因。"
+   }
+  ],
+  "gap_zh": "没有逐帧状态和接触日志，25 fps 画面不能核实毫秒级接触及 0.05 秒失握窗口；但反弹球未到双手便落地，不影响节点未完成的判断。log step 226 是另启的诊断测试，不作本次评分轨迹的逐帧状态；agent 对所有接球方案不可行的推算未经核实，不能据此认定任务无解。"
+ },
+ "summary_zh": "机器人右手投出网球并命中圆靶，评分器记录原生阶段 2/4、偏靶 0.189 米；DAG 要求完整投墙接稳球，因此最终完成 0/1 节点。视频约 2.08–2.36 s 显示反弹球在返回双手前落地，接球接触从未建立；交付脚本只安排投掷而未安排反弹承接，支持跨阶段规划不足。没有逐帧接触状态，细微失握窗口无法核实。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点未达成、断点 wall_throw_caught、没建立接触与推理/规划根因均与视频、评分器 stages 2/4 和 log 221/226/227 一致；未发现需改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:48:24Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--ffybj5m/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--ffybj5m/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "03d2f5b91a8d9622b11fb71416ea3221148a12cca973c29394a8d239da72de30",
+   "video_sha256": "424b0ba4258a6c3955aabb60e76ff2b2373667e1414e455bfa2ea7d14e410ab8",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--is3atbu.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--is3atbu.json
new file mode 100644
index 0000000000000000000000000000000000000000..20ee5912544f94641b4af9ab19376677aa3d30ad
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--is3atbu.json
@@ -0,0 +1,195 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-privileged--is3atbu",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 324,
+   "sim_steps": 324,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 4,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 6.63,
+   "bad_trajectory": 0,
+   "metric_catch_height_m": 0.77,
+   "metric_release_speed_m_s": 16.42,
+   "metric_target_offset_m": 0.001,
+   "metric_wall_speed_m_s": 14.69,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 194,
+   "agent_resets": 7752,
+   "agent_replays": 38,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-privileged--is3atbu/replay.mp4",
+  "duration_s": 3.28,
+  "fps": 25.0,
+  "frames": 82,
+  "sha256": "6a777ac4a01581eb4b31e4050beb476680fdcd289562aacb7fb1d4ee8d5bf66e"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.24,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 1.36,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–1.28 s，黄绿色球随机器人的右手向后、向下摆动；视频 1.32–1.60 s，球与右手分离，向墙面圆靶飞去。机器人以双脚支撑地面。"
+     },
+     {
+      "t": 1.68,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 1.60–1.76 s，球到达红白圆靶附近后反向飞回；视频 1.80–2.12 s，球沿下降轨迹接近正在屈膝下蹲的机器人。"
+     },
+     {
+      "t": 2.28,
+      "log_step": 251,
+      "text_zh": "观察：视频 2.12–2.20 s，球在机器人左手所在的躯干侧方减速，视频 2.28–2.36 s 可见球随该手短距离移动，之后被躯干遮挡；右手仍在头部附近。log step 251 的最终生成器裁判输出记录 catch 为左手，发生于仿真 2.232 s。"
+     },
+     {
+      "t": 3.24,
+      "log_step": 251,
+      "text_zh": "推断：视频 2.40–3.24 s 的逐帧画面显示机器人保持屈膝及接球姿态，未见球离开手边后落地；最后的球和指掌接触被躯干遮挡。评分器 reward 的 success=1、replay_success=1、deterministic=1、stages_reached=4/4，且 log step 251 的最终交付验证记录 324 行动作、仿真 3.24 s 完成 hold 与 success；结合末帧视频 3.24 s，判定完整投墙接稳球节点在视频约 3.24 s 达成并保持至结束。此时间以视频末帧作锚点，并非由 agent 工作时间换算。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.00–1.28 s，机器人右手握球摆臂后投出；视频 2.28–2.36 s，反弹球随左手移动，右手没有同时接球。推断：结合 log step 251 的最终裁判左手 catch 记录和评分器四阶段成功结果，接球为左手指掌包握；右手投球与左手接球依次发生，掌侧接触细部受遮挡。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–3.24 s，机器人双脚接触场地支撑身体，投出后屈膝下蹲配合接球，至视频 3.24 s 末帧仍由脚部支撑。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": 0.001
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 16.42
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 14.69
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": 0.77
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人完成右手投球、命中圆靶、左手接回并持稳的连续动作，唯一节点“投墙接稳球”在视频约 3.24 s 达成，最终为 1/1，评分器成功且确定性回放成功。视频 2.28–2.36 s 可见反弹球随左手移动，随后机器人保持下蹲接球姿态至视频 3.24 s。没有状态日志，左手与球的末段接触受躯干遮挡，无法仅凭 25 帧/秒画面独立核实指掌几何及 0.05 秒失握容差，因此完整持球验收结合评分器四阶段成功结果判定。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "唯一节点“投墙接稳球”在视频 3.24 s 达成与评分器 success、4/4 阶段一致；接球为左手、脚支撑的接触标注合理，时间锚点（仿真 2.23 s 接住 +1 s hold ≈ 3.23 s）自洽，无失败层。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:44:32Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--is3atbu/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--is3atbu/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "b048d428631534ac2b6c7e4f60935e4cfe82d6e720de5240f7362980c6c4dc01",
+   "video_sha256": "6a777ac4a01581eb4b31e4050beb476680fdcd289562aacb7fb1d4ee8d5bf66e",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--jfby59x.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--jfby59x.json
new file mode 100644
index 0000000000000000000000000000000000000000..e4ae7ee956f15bad548d063c8a470887789721fc
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--jfby59x.json
@@ -0,0 +1,195 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-privileged--jfby59x",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 1,
+  "reward": 1.0,
+  "source": "catalog",
+  "raw": {
+   "success": 1,
+   "replay_success": 1,
+   "deterministic": 1,
+   "n_actions": 368,
+   "sim_steps": 368,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 1.0,
+   "stages_reached": 4,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 6.48,
+   "bad_trajectory": 0,
+   "metric_catch_height_m": 0.76,
+   "metric_release_speed_m_s": 18.67,
+   "metric_target_offset_m": 0.073,
+   "metric_wall_speed_m_s": 16.85,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 1219,
+   "agent_resets": 0,
+   "agent_replays": 2,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-privileged--jfby59x/replay.mp4",
+  "duration_s": 3.72,
+  "fps": 25.0,
+  "frames": 93,
+  "sha256": "fd2261044e83209659217b4e4977194a824ee1a16ef410063008f6b98fa35e8e"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 1,
+  "final": 1,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [
+     [
+      3.68,
+      null
+     ]
+    ],
+    "final": true,
+    "evidence": [
+     {
+      "t": 1.92,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–1.80 s，机器人在双脚支撑下将右手持球向后摆动蓄力；视频 1.92–1.96 s，黄绿色球离开向前挥出的右手并飞向墙面。"
+     },
+     {
+      "t": 2.2,
+      "log_step": null,
+      "text_zh": "观察：视频 2.00–2.16 s，球向红白圆靶飞近；视频 2.20–2.28 s，球在圆靶区域折返，视频 2.32–2.60 s 向机器人飞回并下降到腰腿附近。"
+     },
+     {
+      "t": 3.68,
+      "log_step": 93,
+      "text_zh": "观察：视频 2.64–2.80 s，回球进入机器人手边，部分被躯干遮住；视频 3.08–3.68 s，机器人将接球手抬至头部附近，球随手停留，双脚仍支撑地面。log step 93 的官方验证器对交付的 368 行轨迹报告 success=true、failed=null、release→target→catch→hold 全部完成，仿真终点 time_s=3.68；log step 95 再次回放成功且终态哈希一致。"
+     },
+     {
+      "t": 3.68,
+      "log_step": 93,
+      "text_zh": "推断：上述视频中的投出、靶区折返及接回后随手停留，与评分器 success=1、stages_reached=4/4、deterministic=1 一致，故“投墙接稳球”完成并保持为真。完成锚点取视频最后一帧 3.68 s；log step 93 的交付回放终点仿真 3.68 s 与此吻合，未将 agent 工作时间换算为视频时间。合法持握的细小接触几何和短暂失握容差由成功判定补足。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：机器人先用右手包住网球并挥臂投出，随后单手接回并抬起持球。推断：结合视频 2.64–3.68 s 的接回及抬手过程、log step 92 的交付生成脚本左手闭合设置和 log step 93 的官方 catch/hold 成功结果，接球为左手手指与掌部包住网球；右手投出与左手接住先后发生。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–3.68 s，机器人双脚接触地面，腿和躯干在投球、接球及抬手持球期间维持站立支撑。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": 0.073
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 18.67
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 16.85
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": 0.76
+  }
+ ],
+ "failure": null,
+ "summary_zh": "机器人完成唯一节点“投墙接稳球”：视频约 1.92 s 右手投出，约 2.20 s 球在圆靶处折返，接回后随左手抬起并停留，完成锚点为视频约 3.68 s。官方交付回放记录 release、target、catch、hold 四阶段全部完成，二次回放成功且终态哈希一致，评分器记录撞靶偏差 0.073 米。无状态日志，接球瞬间部分被躯干遮挡，25 帧/秒的视频不能独立核实掌侧几何及 0.05 秒失握容差，相关验收依据官方成功判定。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "单节点 wall_throw_caught 成功，视频与官方回放（log step 93/95）一致；时间锚点 3.68 s 在视频内，接触标注（抓握、腿/躯干支撑）符合所见，无失败层，未发现需改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:50:23Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--jfby59x/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--jfby59x/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "d16e589e49c165835935171ed2623a37340e9bb2e65fe20e14efe199a75d1d7b",
+   "video_sha256": "fd2261044e83209659217b4e4977194a824ee1a16ef410063008f6b98fa35e8e",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--jnpvxdv.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--jnpvxdv.json
new file mode 100644
index 0000000000000000000000000000000000000000..9e2637f86f88e649480bcfd92a3048633170bcf6
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--jnpvxdv.json
@@ -0,0 +1,220 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-privileged--jnpvxdv",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 0,
+  "reward": 0.5,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 1500,
+   "sim_steps": 198,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 0.5,
+   "stages_reached": 2,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 4.74,
+   "bad_trajectory": 0,
+   "metric_release_speed_m_s": 15.1,
+   "metric_target_offset_m": 0.227,
+   "metric_wall_speed_m_s": 13.33,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 175,
+   "agent_resets": 9650,
+   "agent_replays": 7,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 3.477,
+   "peak_processes": 27,
+   "peak_disk_gb": 0.246,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-privileged--jnpvxdv/replay.mp4",
+  "duration_s": 2.0,
+  "fps": 25.0,
+  "frames": 50,
+  "sha256": "fbfd0bf8a38b58a5805ce2d0799379d5ab5b17d109089efc5f150586f440288b"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.0,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.88 s，机器人右手握着网球完成引臂；视频 1.00 s 球已离开右手，随后飞向墙面圆靶。"
+     },
+     {
+      "t": 1.36,
+      "log_step": 100,
+      "text_zh": "观察：视频 1.32–1.40 s，球到达圆靶上缘附近并反向飞回；log step 100 对最终交付文件的官方 CLI 回放报告 stages=[release,target]，target_offset_m=0.227，小于 0.25 米阈值，与评分器 reward 一致。"
+     },
+     {
+      "t": 1.8,
+      "log_step": 100,
+      "text_zh": "观察：视频 1.72–1.88 s，反弹球下降经过机器人前方，从双手下方穿过而未随手停留；视频 1.92–1.96 s 球继续降至近地面，机器人仍站立。log step 100 的交付文件回放以 the ball touched the ground 结束，仅有 release、target，无 catch、hold；完整投墙接稳球事件未完成。"
+     },
+     {
+      "t": null,
+      "log_step": 120,
+      "text_zh": "观察：log step 93 保存并回放该轨迹，log step 100 再生成和回放仍仅达到 release、target；log step 120 比较备份确认 gen.py 与 trajectory.npz 未变。log step 121 明确说明后续扭腰接球试验未更新交付，因此这些试验的接近距离和摔倒不属于本段评分视频。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.48、0.88 s，机器人右手手指围住网球并带球引臂，视频 1.00 s 放球投出；反弹后未观察到任一手重新握球。此处抓握指投出前的实际持球。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–1.96 s，机器人双脚支撑地面，配合右臂投球及随后的抬臂接球动作，未见摔倒。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": 0.227
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 15.1
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 13.33
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "wall_throw_caught",
+   "t": 1.8,
+   "log_step": 100,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 1.72–1.88 s，机器人抬臂迎接反弹球，但球从双手下方通过并继续下落，未被接住；log step 100 对最终交付文件的官方回放确认仅完成 release、target，并以球触地结束。失败发生在投墙接稳球步骤的接球过程中。"
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 1.8,
+    "log_step": null,
+    "evidence_zh": "观察：视频 1.64–1.88 s，反弹球快速向机器人飞回并下降，机器人双手在球路上方，抬臂动作未跟随下降球路；视频 1.80–1.88 s 球从手下穿过，随后视频 1.92–1.96 s 继续落向地面，未形成随手停留的接球。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "source_zh": "视频 / 日志",
+    "evidence_zh": "推断：视频 1.64–1.88 s 的接球动作未将手送到反弹球经过的位置。log step 92–93 对随后保存的同一候选轨迹的测试报告最小手球距离约 0.607 米，官方回放仅有 release、target；log step 100 再次验证最终文件仍在球落地前没有接住。结合视频与交付验证，接球运动控制未实现有效拦截；现有证据不足以进一步区分轨迹生成、跟踪或接触参数问题。"
+   }
+  ],
+  "gap_zh": "没有逐帧状态或接触日志；25 帧/秒的远景视频不能核实毫秒级擦碰及精确掌侧几何。视频和最终交付文件的官方回放足以确定未接住并完成持稳，但不能验证 agent 关于站姿接球不可达的判断，也不能确定控制失败的更细子类；未将后续未交付试验中的摔倒或较小手球距离计入本次轨迹。",
+  "unmet": {
+   "wall_throw_caught": false
+  }
+ },
+ "summary_zh": "机器人右手投出网球并命中圆靶，评分器记录释放、命中两阶段及 0.227 米靶心偏差，但唯一语义节点“投墙接稳球”未完成，最终为 0/1。视频约 1.80 s，反弹球从双手下方穿过并继续落向地面，接球动作未实现有效拦截；最终交付回放也仅有 release、target。没有状态与接触日志，控制失败的细分机制仍不确定，后续未交付试验不计入本次进度或失败现象。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 8,
+    "summary_zh": "节点未完成、断点、现象（动态时机）和根因（控制，未细分）与视频及交付文件回放一致；球从手下方落地，评分器仅 release、target，grader 0.5/2 阶段与之相符。未发现需改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:46:14Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--jnpvxdv/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--jnpvxdv/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "5f3ad01916e2263f8fe3b3d0d2d0f66368ca905e349c9b87347c441382817eb4",
+   "video_sha256": "fbfd0bf8a38b58a5805ce2d0799379d5ab5b17d109089efc5f150586f440288b",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--s2nz7hi.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--s2nz7hi.json
new file mode 100644
index 0000000000000000000000000000000000000000..f1b943235e38c89b3319955a95397c491ce1d908
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--s2nz7hi.json
@@ -0,0 +1,205 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-privileged--s2nz7hi",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 0,
+  "reward": 0.5,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 130,
+   "sim_steps": 130,
+   "horizon": 1500,
+   "episode_over": 0,
+   "progress": 0.5,
+   "stages_reached": 2,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 3.24,
+   "bad_trajectory": 0,
+   "metric_release_speed_m_s": 18.42,
+   "metric_target_offset_m": 0.12,
+   "metric_wall_speed_m_s": 16.71,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 42,
+   "agent_resets": 135,
+   "agent_replays": 0,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.547,
+   "peak_processes": 9,
+   "peak_disk_gb": 0.252,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-privileged--s2nz7hi/replay.mp4",
+  "duration_s": 1.32,
+  "fps": 25.0,
+  "frames": 33,
+  "sha256": "271b1d7d5f7f6d2f445edd5668b1c84bc34a552fe1bf51160e037dfdcf24a49e"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.4,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.40 s，机器人右手携黄绿色网球向后摆臂蓄力，双脚在地面支撑；视频约 0.64 s 球已离开右手飞向墙面。初始握球不满足投墙后接稳的完整事件。"
+     },
+     {
+      "t": 0.92,
+      "log_step": 63,
+      "text_zh": "观察：视频约 0.88–0.96 s，球到达红白圆靶附近后反向飞回；保存轨迹的执行结果（log step 63）报告 stages=[release,target]、target_offset_m=0.12，与评分器的 2/4 阶段和 0.12 米偏差一致。"
+     },
+     {
+      "t": 1.28,
+      "log_step": 63,
+      "text_zh": "观察：逐帧检查视频 0.80–1.28 s，反弹球向机器人飞回并下降，到末帧仍未显示被手包住后随手停留；机器人一手仍举在头旁，另一手在腰腹附近。log step 62 的脚本只执行蓄力、投出和投后维持姿态，随后保存；log step 63 报告 steps=130、done=False、success=False，评分器 episode_over=0、stages_reached=2，没有 catch 或 hold。"
+     },
+     {
+      "t": null,
+      "log_step": 63,
+      "text_zh": "推断：交付片段到此尚未完成反弹接球及接后满 1 秒持球，完整事件从未成立，因此该节点 intervals=[]、final=false；原生 2/4 阶段进展不等于完成一个 DAG 节点。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.00–0.40 s，右手手指包住网球，球随右手向后摆动，随后由右手投出；视频结束前未见任何一只手完成反弹后的包握接球。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–1.28 s，机器人双脚接触场地，腿部承重支撑投球和投后转体；末段身体倾斜，但片段内未见非脚部触地或摔倒。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": 0.12
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 18.42
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 16.71
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "wall_throw_caught",
+   "t": null,
+   "log_step": 63,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 0.00–1.28 s 的机器人已开始投墙接球并命中圆靶，但交付片段在球返回、尚未显示接稳时结束。log step 62 保存的脚本止于投后维持姿态，log step 63 的保存结果为 130 步、release/target 两阶段、done=False；评分器也报告 episode_over=0。可定位未完成的是投墙接稳球步骤，但无法定位片段内一个已发生的不可逆接球失败时刻，故 t=null。"
+  },
+  "unmet": {
+   "wall_throw_caught": false
+  },
+  "phenomena": [],
+  "root_causes": [],
+  "gap_zh": "观察：inputs/ 内无状态日志；视频仅覆盖交付的 130 步部分轨迹，末帧为视频 1.28 s，未覆盖完整接球机会及接后 1 秒验收。会话末尾 log step 92 明确记录模型提供方过滤中断，且此前仍在开发接球控制；没有最终完成报告。log step 64 中球在仿真 1.474 s 落地来自另一次执行 80 步投后控制的延长测试，交付脚本的投后控制仅 60 步，该落地不在视频内，不能用作本轨迹的失败现象。推断：保留评分器 success=0 及节点未完成；已保存部分轨迹，但没有片段内可确认的接球错失、接后脱手或不可逆判负，故现象留空；过滤前也无足够证据区分尚未完成开发与某种导致本次失败的控制、认知或交付错误，故根因留空。"
+ },
+ "summary_zh": "机器人右手投出网球并命中圆靶，评分器记录 release、target 两阶段（2/4，偏差 0.12 米），但交付视频结束前未接稳，唯一 DAG 节点完成 0/1。断点定位为操作中的投墙接稳球；交付文件仅保存 130 步，保存时仍 done=False，随后会话在继续开发时遭模型过滤中断（log step 92）。视频未覆盖完整接球和持球验收，后续延长测试中的落地不属于交付片段，因此不臆定失败现象或根因。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "节点未完成、断点（投墙接稳球，操作中）与证据一致：视频仅1.32 s，末帧球仍在返回途中，交付脚本（log id 62）不含接球动作，评分器 stages=2/4、episode_over=0。现象/根因留空并写 gap_zh 符合 CASES FAIL-23（交付片段不含接球尝试，不能命名接球类现象），未发现需要改动的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:52:08Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--s2nz7hi/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--s2nz7hi/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "f0f9e9c5b4ae7398133e083f7d1592f9412ffbb1ee8dadaf63810742ab5c9ed1",
+   "video_sha256": "271b1d7d5f7f6d2f445edd5668b1c84bc34a552fe1bf51160e037dfdcf24a49e",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--zxutnbv.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--zxutnbv.json
new file mode 100644
index 0000000000000000000000000000000000000000..319c91cf039c1baf09a27467b498733c51addd09
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-privileged--zxutnbv.json
@@ -0,0 +1,210 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-privileged--zxutnbv",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 0,
+  "reward": 0.5,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "replay_success": 0,
+   "deterministic": 1,
+   "n_actions": 246,
+   "sim_steps": 246,
+   "horizon": 1500,
+   "episode_over": 1,
+   "progress": 0.5,
+   "stages_reached": 2,
+   "stages_total": 4,
+   "fell": 0,
+   "replay_s": 5.28,
+   "bad_trajectory": 0,
+   "metric_release_speed_m_s": 16.83,
+   "metric_target_offset_m": 0.2,
+   "metric_wall_speed_m_s": 15.11,
+   "grader_error": 0,
+   "video_rendered": 1,
+   "agent_sessions": 132,
+   "agent_resets": 8780,
+   "agent_replays": 5,
+   "agent_cost_self_reported": 1,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.895,
+   "peak_processes": 34,
+   "peak_disk_gb": 0.983,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0,
+   "session_log_tampered": 0,
+   "session_log_from_monitor": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-privileged--zxutnbv/replay.mp4",
+  "duration_s": 2.48,
+  "fps": 25.0,
+  "frames": 62,
+  "sha256": "353d65a7ad578a8ee57e635b4d27a36d6620d67851b72c6fd2a727f808b6089c"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 1.52,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–1.48 s，机器人右手持黄绿色网球蓄力；视频 1.52 s 球已离手，视频 1.84–1.88 s 球到达灰墙圆靶附近并反向飞回。"
+     },
+     {
+      "t": 2.32,
+      "log_step": 354,
+      "text_zh": "观察：视频 2.24–2.44 s，反弹球穿过机器人手旁后继续向后下方飞行，没有被手包住并随手停留；视频 2.40–2.44 s 球已接近地面。log step 354 对最终交付文件的回放裁判输出 stages=[release,target]、success=false、failed=the ball touched the ground，与画面一致；完整投墙接稳球事件未完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.00–1.48 s，机器人右手握着网球随臂蓄力，视频 1.36 s 可见球位于合拢的右手内，随后右手松开投出；反弹后的左手接球动作没有形成可见持握，不把它记作已发生的抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–2.44 s，机器人以脚部接触地面支撑投接动作；后段屈膝并抬起一脚，另一脚仍支撑身体。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": 0.2
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 16.83
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": 15.11
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "wall_throw_caught",
+   "t": 2.32,
+   "log_step": 354,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 2.28–2.36 s，机器人在尝试承接回球时未将手合到球上，球从手旁掠过并继续下降，之后没有修复；log step 354 的最终交付回放裁判仅记录 release、target 两阶段，并以球触地结束。断点为投墙接稳球步骤中的接球动作。"
+  },
+  "unmet": {
+   "wall_throw_caught": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 2.32,
+    "log_step": 354,
+    "evidence_zh": "观察：视频 2.24–2.36 s，机器人双手未跟上反弹球经过的轨迹，球从手旁通过后持续下落，没有停留在任何一只手中；log step 354 的交付回放没有 catch 或 hold 阶段。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "推断：交付轨迹的接球控制未能使左手的承接位置、运动与合手动作匹配回球。log step 272 的交付生成脚本先令左臂到固定 ready 姿态，在球越过指定 x 阈值后切换为固定力矩摆臂，并在球掌距离小于 gd 时才命令闭合；log step 280 保留该脚本并设置最终接球参数，log step 354 再次生成和回放仍未接住。视频 2.24–2.36 s 可见回球经过手旁而手未承接住，支持控制归因；不能据此确认是可达性、跟踪还是手指闭合速度单独导致。"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志；25 fps 的远景无法分辨瞬时擦碰、具体手指接触和掌侧几何，也不足以独立核实关节可达性或合手延迟。因此根因只标控制，不采用 agent 关于“够不着”或后续下蹲测试的结论作为交付轨迹事实。完整接球节点未达到由可见球路和最终交付回放裁判共同确定。"
+ },
+ "summary_zh": "机器人右手投出网球并命中圆靶，原生裁判完成 release、target 两阶段，但完整“投墙接稳球”节点为 0/1。视频约 2.32 s，反弹球从手旁掠过后下落，交付回放裁判以球触地结束；接球控制未能匹配回球，现象为动态时机。缺少逐帧接触和关节状态，不能进一步区分可达性、跟踪或合手速度原因。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点未达成（0/1）、断点 wall_throw_caught、现象动态时机（CASES FAIL-14）、根因控制均与视频和 log step 354 的交付回放裁判（stages=[release,target]，球触地）一致；未发现需要改动的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:55:39Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--zxutnbv/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-privileged--zxutnbv/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "e0ff6824f2318734f6321f93179527015ea1ea63f4ba0901f049b3fa7225f0bd",
+   "video_sha256": "353d65a7ad578a8ee57e635b4d27a36d6620d67851b72c6fd2a727f808b6089c",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--bdoy9nv.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--bdoy9nv.json
new file mode 100644
index 0000000000000000000000000000000000000000..9294eb750de94ac48abb88c0b82365675272ecdd
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--bdoy9nv.json
@@ -0,0 +1,212 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-standard--bdoy9nv",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 138,
+   "sim_steps": 723,
+   "calls": 19,
+   "rows": 723,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 907.0,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 1.007,
+   "peak_processes": 10,
+   "peak_disk_gb": 1.316,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-standard--bdoy9nv/replay.mp4",
+  "duration_s": 1.4,
+  "fps": 25.0,
+  "frames": 35,
+  "sha256": "3e5d003250827a23cf825c15ecf40a6766ed3902884e3f2da37af87840855702"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.64,
+      "log_step": 145,
+      "text_zh": "观察：视频 0.00–0.56 s，机器人右手携初始黄绿色网球后摆并前挥；视频约 0.64 s 起球与右手分离，随后向墙前飞。机器人双脚始终支撑地面。"
+     },
+     {
+      "t": 1.36,
+      "log_step": 149,
+      "text_zh": "观察：视频 0.64–1.36 s，球沿低弧线前飞后下降到墙前地面，没有在红白圆靶处命中并反向返回，也没有被任一手接住。log step 149 记录最后一次重置后的提交，机器人服务返回“the ball touched the ground”、stages_reached=1 of 4、executed_rows=138，与这段视频的落地结局一致；完整投墙接稳球节点从未成立。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.00–0.56 s，右手手指围住初始网球并带球后摆、前挥，随后张手投出；左手摆动但未与球建立共同持握，未发生反弹接球。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–1.36 s，机器人双脚接触地面并支撑身体，配合右臂投球，没有摔倒。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "wall_throw_caught",
+   "t": 0.64,
+   "log_step": 149,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 0.64 s，机器人右手将球投出；球后续沿低弧线飞向墙前并下降，视频 1.36 s 已接近地面，未完成圆靶命中或回弹接球。log step 149 的最终提交由服务以球触地判负，原生阶段仅完成 1/4；断点属于投墙接稳球步骤中的投掷阶段。"
+  },
+  "unmet": {
+   "wall_throw_caught": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 0.64,
+    "log_step": 149,
+    "evidence_zh": "观察：视频 0.64–1.36 s，机器人投出的网球没有飞到红白圆靶并回弹，而是沿低弧线落向墙前地面；机器人仍站立。log step 149 的服务结果为球触地且仅完成释放阶段，说明本次投掷没有形成所需的命中轨迹。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 0.64,
+    "log_step": 145,
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "推断：log step 145 的最终脚本以固定肩、肘、腕角度序列前挥并按固定序号张开右手，随后一次性提交整段动作；目标明确为投中圆靶后接住。视频 0.64–1.36 s 的实际球路却在命中前落地，log step 149 的服务判负与之相符，支持投掷动作与释放配合的控制未达到要求；没有足够证据区分出手速度、方向、接触滑移或关节跟踪中的具体原因。"
+   }
+  ],
+  "gap_zh": "没有状态日志，也没有本次最终投掷的球速、释放瞬间手球接触和关节跟踪记录；网球在远景中很小，无法精确分离出手角度、速度与手部接触对低球路的贡献，因此根因仅标控制。四项原生诊断指标在 run.json 的评分字段中均未报告，值留 null。"
+ },
+ "summary_zh": "评分尝试中，机器人右手持球后摆并投出，但球沿低弧线飞行后落地，未完成投墙接稳球，节点终态为 0/1；log step 149 的服务报告原生阶段仅达 1/4。断点在视频约 0.64 s 的投掷阶段，现象为位姿不准，固定投掷动作与释放配合未产生合格球路，根因标控制。缺少状态与释放运动学记录，不能进一步确定具体控制子因。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "视频 0.56–1.36 s 球从右手飞出，沿低弧线向墙飞去后落地，没有命中圆靶或被接住；节点 false、断点在 wall_throw_caught、位姿不准加控制的标注与 log step 149 的服务结果（球触地、1/4 阶段）一致，未发现需要修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:45:08Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--bdoy9nv/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--bdoy9nv/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "c892193c88d2b84655124209e914ea4225793486a100be2d2dbfc1804447da4e",
+   "video_sha256": "3e5d003250827a23cf825c15ecf40a6766ed3902884e3f2da37af87840855702",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--clmmtwi.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--clmmtwi.json
new file mode 100644
index 0000000000000000000000000000000000000000..f803339b5617c7be00e848fb47b03dcb17fe4e35
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--clmmtwi.json
@@ -0,0 +1,211 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-standard--clmmtwi",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 320,
+   "sim_steps": 1676,
+   "calls": 19,
+   "rows": 1266,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 6744.7,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0,
+   "peak_memory_gb": 13.701,
+   "peak_processes": 33,
+   "peak_disk_gb": 3.032,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-standard--clmmtwi/replay.mp4",
+  "duration_s": 3.24,
+  "fps": 25.0,
+  "frames": 81,
+  "sha256": "63984b14505b683489e03bc899ba66ca818a1d7be39677f648bc93c09b694b89"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.8,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–2.80 s，机器人双脚站立，右臂由身前向后摆并停留；没有网球飞向墙靶、撞墙反弹或被手重新接住的过程。初始握球不是完成投墙接球事件。"
+     },
+     {
+      "t": 3.04,
+      "log_step": 166,
+      "text_zh": "观察：视频 3.00–3.20 s，黄绿色网球从右手附近向下运动，在视频 3.04 s 已与右手分离，随后落至机器人脚边；右臂继续向前上方空摆，网球没有到达墙靶。log step 166 中机器人服务在重置到 episode 6 后返回 stages_reached=1、stages_total=4、outcome=\"the ball touched the ground\"，与本视频的落球终局一致；只完成释放阶段，完整节点未达。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "推断：任务初态说明网球由闭合右手握住，视频 0.00–2.96 s 右手合拢并随右臂后摆，视频 3.00–3.04 s 球从该手附近掉出；实际接触是右手握持初始网球，未发生反弹后的接球。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–3.20 s，机器人双脚接触场地支撑站立，配合右臂后摆和向前挥动，末尾仍站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "wall_throw_caught",
+   "t": 3.04,
+   "log_step": 166,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 3.04 s，机器人正在将后摆的右臂向前挥动，网球已离开右手向脚边下落；视频 3.20 s 球在地面附近，未撞墙或接回。log step 166 的第 6 局服务结果为球触地、原生阶段 1/4，投墙接稳球步骤未完成且未修复。"
+  },
+  "phenomena": [
+   {
+    "cat": "接触",
+    "item": "没保持住",
+    "t": 3.04,
+    "log_step": 166,
+    "evidence_zh": "观察：视频 2.92–3.04 s，机器人右手从后摆位置启动前挥，黄绿色球从手部附近分离并下降；视频 3.08–3.20 s 球继续落到脚边，右手则空摆至头旁。推断：结合 log step 150 的 throw_rows 闭合指令逻辑和 log step 165 的最终摆臂配置，球在计划开手的摆臂后段之前已脱离，属于原有握持未保持住；log step 166 的服务球触地结果佐证终局。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "抓取和接触设计",
+    "alt_feasible": "未知",
+    "source_zh": "日志",
+    "evidence_zh": "推断：log step 150 的 throw_rows 在摆臂前段仍给右手闭合命令，log step 165 保留这一握持逻辑并修改最终摆臂参数；视频 2.92–3.04 s 却显示球在前挥刚启动时掉出，说明该右手握持方式与摆臂负载的组合未能保持球掌接触。log step 167 的本局手部读数随摆臂变化可作辅助，但不把 agent 对其的解释当作接触力测量；具体摩擦、指位和负载分量尚不能区分。"
+   }
+  ],
+  "gap_zh": "无逐帧状态或接触力日志，单一场地视角下手指与球的接触细节较小，25 fps 视频无法精确核实毫秒级失握时刻或区分摩擦不足与指位不适配。评分字段没有报告 DAG 的四项原生指标，故均填 null；agent 本地替代模型的出球速度和靶偏差不作为本局实测。agent 在 log step 172 声称可行投接不存在，但其自建模型不能排除所有合法方案，因此不据此归因为任务或环境。",
+  "unmet": {
+   "wall_throw_caught": false
+  }
+ },
+ "summary_zh": "本次评分为第 6 局，机器人完成了初始持球与后摆，但视频约 3.04 s 前挥时球提前脱手并落地，未命中墙靶或反弹接稳，节点最终 0/1；log step 166 的服务结果仅达释放阶段（1/4）。断点为投墙接稳球的操作中，现象是没保持住，根因推断为握持与摆臂负载不匹配。缺少接触与力状态，无法进一步确定具体指位或摩擦问题；成功参考视频仅用于对照完整投接的可见表现。",
+ "conf": 7,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点 wall_throw_caught 未达、断点（操作中，球在前挥启动时脱手落地）、现象“没保持住”与根因“抓取和接触设计”均与视频和 log 一致；未发现需改内容的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:02:03Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--clmmtwi/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--clmmtwi/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "e47b4b12f0ba6379c188a86fd5600f7e2e8df9455ffc8fe02186b1babce3a7e3",
+   "video_sha256": "63984b14505b683489e03bc899ba66ca818a1d7be39677f648bc93c09b694b89",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--fqktfbw.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--fqktfbw.json
new file mode 100644
index 0000000000000000000000000000000000000000..26b5a7cfaed8b9b86981e34949e5565c8cc1a03c
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--fqktfbw.json
@@ -0,0 +1,237 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-standard--fqktfbw",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 125,
+   "sim_steps": 125,
+   "calls": 6,
+   "rows": 110,
+   "resets": 0,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 6941.2,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-standard--fqktfbw/replay.mp4",
+  "duration_s": 1.28,
+  "fps": 25.0,
+  "frames": 32,
+  "sha256": "31bf8a3d907793cb7b04db202c67d930763d59374679542c53b485e42dbb9631"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.64,
+      "log_step": 178,
+      "text_zh": "观察：视频 0.48–0.72 s，机器人右手包住黄绿色网球并向身后下方摆臂，双脚保持站立；log step 178 的实时服务返回同一 episode 1、control_steps=70、holding_right=1，证实该回合已经开始投球准备。"
+     },
+     {
+      "t": 0.88,
+      "log_step": 190,
+      "text_zh": "观察：视频 0.80–0.88 s，机器人右臂向前摆动，球离开右手，随后向灰墙飞行；log step 190 的该回合本体感知采样中，仿真 0.76 s 的 holding_right=1，仿真 0.79 s 已为 0。"
+     },
+     {
+      "t": 1.24,
+      "log_step": 200,
+      "text_zh": "观察：视频 1.12–1.24 s，球靠近墙面，末帧位于红白圆靶下方；全片没有反弹球回到机器人手中并保持的过程。log step 200 的实时服务裁判返回 done=true、outcome=\"the ball hit the wall outside the target\"、stages_reached=1/4、仿真 sim_time_s=1.25；这表示只完成释放，完整投墙接稳球事件未达成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.48–0.72 s，机器人右手的弯曲手指围住黄绿色网球，持球完成后摆；随后右手松开投出。左手未与球接触，未发生反弹后的抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0–1.24 s，机器人双脚支撑地面，配合右臂后摆与向前投球，期间保持站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "wall_throw_caught",
+   "t": 1.24,
+   "log_step": 200,
+   "where": "操作中",
+   "evidence_zh": "观察：机器人正在完成投墙接球动作，视频 0.84–1.24 s 显示球从右手飞向墙面，末帧落点方向低于红白圆靶；log step 200 的实时服务裁判在同一 episode 1 判定球撞墙脱靶并终止，完成阶段为 1/4。该步骤未在后续修复；log step 201–208 只有本地调试，没有新的实时动作或重置。"
+  },
+  "unmet": {
+   "wall_throw_caught": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 1.24,
+    "log_step": 200,
+    "evidence_zh": "观察：视频 1.12–1.24 s，机器人投出的球到达红白圆靶下方，未进入圆靶区域；log step 200 的实时服务裁判明确报告首次撞墙在靶外。目标圆靶固定，失败发生在投出后命中位置偏低，尚未进入反弹接球。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：agent 的投球脚本未将摆臂与开手控制到能命中圆靶的释放状态。log step 189 的脚本预先编排右臂挥动与 grip_right 从 1 降到 0.38、再到 0，log step 190 修正视频帧率参数后执行该脚本，实时采样在仿真 0.79 s 已不持球；视频 0.84–1.24 s 的实际飞行轨迹到达靶下方，log step 200 判脱靶。log step 188 的理想化本地模型命中预测并未在实机服务回合实现，故归为控制，现有证据不足以指定更细子类。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志或原生接触对；视频为 25 fps 的远景，无法精确核实脱手瞬间的接触几何与释放速度，也无法区分开手时序、握球接触变化和摆臂跟踪各自的贡献，故根因仅标控制。评分字段及服务裁判未报告 DAG 四项原生指标的数值，native_metrics 均保留 null。"
+ },
+ "summary_zh": "机器人右手持球后摆并向墙面投出，但视频 1.24 s 的球位于圆靶下方，log step 200 的实时裁判以撞墙脱靶终止回合，只完成释放阶段，唯一节点“投墙接稳球”未达成（0/1）。失败现象为位姿不准，投球脚本的摆臂与开手控制未产生合格投射结果，根因保留为控制。缺少逐帧接触与释放速度记录，无法进一步确定控制子类；回合结束后的本地调试不计入进度。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 7,
+    "summary_zh": "节点、断点、时钟与现象（位姿不准）均有 log step 190/200 与视频支持；但 root_causes[0] 的 item 写成“控制”，不是 taxonomy 中控制类的合法子项（抓取和接触设计/轨迹或逆解/反馈控制/参数调试），需改为具体子项。",
+    "n_issues": 1,
+    "n_major": 1
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点未达成、断点（投墙接稳球，操作中）、现象位姿不准、根因停在控制（03 明文允许子类不清时停在 item 控制）均有视频与 log step 200 支持；时钟均为视频时间，validator 通过。第 1 轮意见已被有证据地驳回，无新证据。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:01:42Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--fqktfbw/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--fqktfbw/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--fqktfbw/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--fqktfbw/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "153f568e4ac1d4fe404da1146fdd8187cd1b37971eacd9d3297e66ee88689f1f",
+   "video_sha256": "31bf8a3d907793cb7b04db202c67d930763d59374679542c53b485e42dbb9631",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--ivybh6r.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--ivybh6r.json
new file mode 100644
index 0000000000000000000000000000000000000000..91ab2855b15e9ba17684c3bc23f04c08ff572b8a
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--ivybh6r.json
@@ -0,0 +1,215 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-standard--ivybh6r",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 375,
+   "sim_steps": 2402,
+   "calls": 241,
+   "rows": 2388,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 6197.0,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.848,
+   "peak_processes": 11,
+   "peak_disk_gb": 1.288,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-standard--ivybh6r/replay.mp4",
+  "duration_s": 3.76,
+  "fps": 25.0,
+  "frames": 94,
+  "sha256": "ed6121984c18df0c5cb3bab083cd8330847a7ea4ac2096cd88de6ec96cd41f74"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.84,
+      "log_step": 161,
+      "text_zh": "观察：视频 2.64–2.80 s 网球仍在机器人右手附近，视频 2.84–2.96 s 网球离开右手向墙面飞行；机器人双脚支撑地面，左手抬起准备接球。"
+     },
+     {
+      "t": 3.16,
+      "log_step": 241,
+      "text_zh": "观察：视频 3.12–3.24 s 网球到达红白圆靶附近后折返；log step 241 从本次最终 episode 6 的裁判结果读取 stages_reached=2、stages_total=4，agent 保存的命令仅选 a6_ 前缀，共 375 行。该结果与视频中的释放及命中过程一致，不能把前五次重置前的尝试计入本节点。"
+     },
+     {
+      "t": 3.56,
+      "log_step": 240,
+      "text_zh": "观察：视频 3.48–3.56 s 反弹球接近并经过左手附近，视频 3.60–3.68 s 球已飞离机器人、向画面左下方下降，视频 3.72 s 未见球停留在任一手中；log step 240 的服务终局字段为 episode=6、outcome=\"the ball touched the ground\"、stages_reached=2。推断：本次未完成反弹接住并持握满 1 秒的完整事件，节点始终未达成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：机器人右手在视频 0–2.80 s 包持黄绿色网球并带球摆臂，随后投出。左手在视频 3.48–3.60 s 迎接反弹球，但未形成可确认的包握；不把这次接球尝试记为成功抓握。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0–3.72 s 机器人以双脚接触地面支撑身体，配合右臂投球和左臂接球，末段仍站立。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "wall_throw_caught",
+   "t": 3.56,
+   "log_step": 227,
+   "where": "操作中",
+   "evidence_zh": "观察：视频 3.48–3.56 s 机器人左手迎向返回的网球，球没有随手停留，视频 3.60–3.68 s 已在机器人侧后方继续飞行下落；log step 227 左手持握读数 hold=0，log step 240 的服务终局结果为球触地、仅达成 2/4 裁判阶段。推断：投墙接稳球步骤在接反弹球时失败，此后追球没有修复。"
+  },
+  "unmet": {
+   "wall_throw_caught": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "动态时机",
+    "t": 3.56,
+    "log_step": 227,
+    "evidence_zh": "观察：视频 3.48–3.56 s 飞行中的网球掠过左手附近，左臂虽迎接并后撤，视频 3.60–3.68 s 球仍独自飞离并下落；log step 227 的 holding 读数为 0，log step 240 裁判未认可 catch 阶段。推断：接球手的运动及闭合未与反弹球到达同步，未建立合法持握；没有证据表明曾接住后失握。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "观察：log step 187 的控制器记录 goal=[0.470,0.426,1.089]、估计 hand=[0.481,0.343,1.061]，手部横向跟踪有约 0.083 米偏差；log step 188 的关节读数亦显示左肩外展与腕俯仰偏离目标。log step 227 服务明确将 left_shoulder_roll_tau 和 left_shoulder_yaw_tau 裁剪到允许范围，且 hold=0；视频 3.48–3.68 s 显示左臂未把反弹球承接住。推断：接球动作执行与跟踪未能实现稳定截获，归为控制；现有证据不足以独立确定接触几何、逆解轨迹或反馈控制中哪一项主导。"
+   }
+  ],
+  "gap_zh": "没有状态日志。视频为单一远景、25 fps，无法分辨视频 3.52–3.56 s 球与左手是否发生瞬时擦碰及具体接触指节，也不能精确核验掌侧几何或 0.05 秒失握窗口；以裁判未达 catch/hold 和球继续飞离共同判定未完成。日志中的手部世界位置是 agent 自身模型估计，扭矩裁剪是服务反馈，二者不足以唯一细分控制根因。评分器 reward 未报告 DAG 的四项原生指标，均保留 null。"
+ },
+ "summary_zh": "本次评分对应最后的 episode 6：机器人右手投球命中圆靶，但左手未接住反弹球，完整“投墙接稳球”节点最终为 0/1。断点约在视频 3.56 s 的接球过程中，现象为动态时机，控制跟踪偏差及服务扭矩裁剪支持控制根因，裁判随后以球触地结束并报告 2/4 阶段。单一远景不能确认瞬时手球擦碰，缺少状态日志，控制根因不再细分。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "节点未达成（最终 0/1）、断点 wall_throw_caught·操作中、动态时机现象与控制根因均与 run.json 和视频一致；episode 6 终局 2/4 阶段（释放、命中）与球触地得到 log step 241 印证。根因细分证据偏弱，但标注已如实说明。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:00:00Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--ivybh6r/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--ivybh6r/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "ef087d37255ea9ba86c0c93f9da77c0a193ea90143fdd81e2e84044ab715731f",
+   "video_sha256": "ed6121984c18df0c5cb3bab083cd8330847a7ea4ac2096cd88de6ec96cd41f74",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--m7mznz5.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--m7mznz5.json
new file mode 100644
index 0000000000000000000000000000000000000000..7945b26db5cd0e971a8789e2d755c1ee2ad55304
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--m7mznz5.json
@@ -0,0 +1,212 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-standard--m7mznz5",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 211,
+   "sim_steps": 1226,
+   "calls": 14,
+   "rows": 783,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 1715.2,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 1,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 0.0,
+   "peak_processes": 0.0,
+   "peak_disk_gb": 0.0,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-standard--m7mznz5/replay.mp4",
+  "duration_s": 2.12,
+  "fps": 25.0,
+  "frames": 53,
+  "sha256": "02719ea06e25cec4753b087d706799547fe6dd922d07f53cea505a4be7488932"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.0,
+      "log_step": 154,
+      "text_zh": "观察：视频 0.00–1.20 s，机器人双脚站立，右臂先向下收后向前挥出；视频 1.28 s 黄绿色球已与右手分离。log step 154 记录最后一次 reset 后为 episode 6，随后提交 143 行最终挥臂动作；此前各局的动作不计入本视频进度。"
+     },
+     {
+      "t": 2.08,
+      "log_step": 156,
+      "text_zh": "观察：逐帧查看视频 1.28–2.08 s，球向墙方向飞出，随后持续下降，在到达圆靶前落到蓝色场地，未发生墙面反弹或手部接回。log step 156 的机器人服务（执行动作并返回裁判结果的 API）对 episode 6 返回 done=true、outcome=\"the ball touched the ground\"、stages_reached=1；与视频只完成出球一致，完整投墙接稳球事件从未完成。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "推断：任务初态明确网球握在右手，视频 0.00–1.20 s 右手随挥臂运动，视频约 1.28 s 球从右手处飞离；右手灵巧手原先包握网球，随后释放。未观察到接回球，左手未与球协作。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–2.08 s 机器人双脚持续接触场地并支撑站立，配合右臂投球，未见摔倒。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "wall_throw_caught",
+   "t": 1.28,
+   "log_step": 154,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 1.28 s，机器人挥动右臂使球离手；视频 1.28–2.08 s 球的飞行在墙前下落结束，没有命中圆靶、反弹或接稳。log step 154 的最终动作与 log step 156 的 episode 6 球触地判负相对应，失败发生于投墙接稳球的投出阶段。"
+  },
+  "unmet": {
+   "wall_throw_caught": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 1.28,
+    "log_step": 156,
+    "evidence_zh": "观察：视频 1.28–2.08 s，右手投出的球先向前飞行后下落，落点仍在墙前，未到达墙上固定红白圆靶；没有形成可接回的反弹轨迹。log step 156 中服务裁判报告球触地、仅达 1 个阶段，与投掷未命中一致。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "t": 1.28,
+    "log_step": 154,
+    "source_zh": "日志 / 视频",
+    "evidence_zh": "推断：log step 154 的最终脚本用预定右肩俯仰加速、右肘俯仰及随后张手完成投掷；log step 156 继续保持最后姿态。视频约 1.28 s 的实际出球及视频 2.08 s 的墙前落地表明，这组挥臂和释放控制未产生到达圆靶所需的出球状态，归为控制问题；没有足够证据进一步确定接触设计、跟踪或参数子类。"
+   }
+  ],
+  "gap_zh": "无逐帧状态日志，远景视频无法精确量化出手速度、角度、手指接触和握力。log step 163 中 agent 将失败归咎于握力上限并声称接球不可达，这些属于 agent 自述，未由本局传感记录验证，不能据此认定握力设计问题或任务不可解；评分字段亦未报告 DAG 中四项原生指标，均记 null。"
+ },
+ "summary_zh": "评分局为最后重置后的 episode 6：机器人站立挥动右臂，球于视频约 1.28 s 离手，随后在墙前落地；未命中、未反弹接回，唯一节点“投墙接稳球”未完成，终态 0/1。断点在该节点的投出阶段，现象为位姿不准，最终脚本的挥臂及释放控制未形成所需球路。缺少本局逐帧接触与出手状态，不能验证 agent 关于握力上限和任务不可达的归因。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "标注与视频、日志一致：评分局为最后一次 reset 后的 episode 6，球约视频 1.28 s 离手，在墙前落到蓝色场地，未命中圆靶、未接回，唯一节点未完成，断点、现象、根因均有证据支持。未发现需修改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:47:49Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--m7mznz5/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--m7mznz5/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "f76163356d891de9ac819fcae3030dfb5145398d7c704556e011120f017b61ff",
+   "video_sha256": "02719ea06e25cec4753b087d706799547fe6dd922d07f53cea505a4be7488932",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--qdgxvtr.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--qdgxvtr.json
new file mode 100644
index 0000000000000000000000000000000000000000..d9fc28fc855e2a71ad9b48cb4d08d10ae5b46b0d
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--qdgxvtr.json
@@ -0,0 +1,232 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-standard--qdgxvtr",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 335,
+   "sim_steps": 1040,
+   "calls": 145,
+   "rows": 540,
+   "resets": 2,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 709.4,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.664,
+   "peak_processes": 10,
+   "peak_disk_gb": 0.329,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-standard--qdgxvtr/replay.mp4",
+  "duration_s": 3.36,
+  "fps": 25.0,
+  "frames": 84,
+  "sha256": "d1740bed72b158f4a05fbd65174f7ee30e119dbd559bc8e7c990eb4f973b2ec7"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 2.4,
+      "log_step": 42,
+      "text_zh": "观察：视频 0.00–2.32 s 机器人右臂持球后摆并前挥，视频 2.36–2.84 s 黄绿色球离开右手飞向墙上圆靶；视频约 2.88 s 球到达圆靶内，随后反向飞回。log step 42 记录 reset 3 后运行 a3，机器人服务（实时仿真与裁判接口）返回 stages_reached=2、stages_total=4，与释放、命中圆靶的画面一致；前两次测试不计入本次节点进度。"
+     },
+     {
+      "t": 3.12,
+      "log_step": 42,
+      "text_zh": "观察：视频 2.92–3.32 s 反弹球在墙前、圆靶左下方持续下降，球与机器人手部之间始终有明显间隔，未到达其接球位置；机器人站在原处，右手停在身体侧前方低处、手掌张开，未向反弹球移动。log step 42 的本次服务终局为 done=True、reason=failed、outcome=\"the ball touched the ground\"，只完成四个验收阶段中的前两个，未接住并持球满 1 秒。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "推断：结合任务初态的右手闭合持球说明、视频 0.00–2.32 s 球随右手移动及随后离手的过程，右手五指在投出前握持网球，之后张开释放；没有反弹后的接球抓握，不标双臂协作。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–3.32 s 机器人以双脚支撑地面完成右臂后摆和前挥，反弹期间仍站在白线后的原处。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "wall_throw_caught",
+   "t": 3.12,
+   "log_step": 42,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 2.88 s 球命中圆靶后反弹；视频 2.92–3.32 s 球在墙前、圆靶左下方下落，与机器人手部保持明显间隔。机器人站在原处，右手停在身体侧前方低处，没有向球移动或包住球。log step 42 中 a3 是最后一次重置后的评分回合，服务报告仅完成 release、target 两阶段，并因球触地失败；失败步骤为完整投墙接稳球。"
+  },
+  "unmet": {
+   "wall_throw_caught": false
+  },
+  "phenomena": [
+   {
+    "cat": "接触",
+    "item": "没建立接触",
+    "t": 3.12,
+    "log_step": 42,
+    "evidence_zh": "观察：视频 2.92–3.32 s 反弹球在墙前向下落，与机器人双手始终有明显间隔，手未接近反弹球，也未出现反弹后手指与手掌共同包球或球随手停留；log step 42 的服务终局报告球触地、stages_reached=2/4。这里指未建立反弹后的接球接触，投出前的右手持球不算接住。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "认知",
+    "item": "推理/规划",
+    "evidence_zh": "推断：本次策略只安排命中圆靶，没有安排能衔接反弹的接球动作。log step 29 的 throw.py 在挥臂循环中把左手闭合指令固定为 0、右手在 k_open 后设为 0，之后只调用 get_obs 等待；log step 42 以该脚本运行最终 a3，并在执行前的笔记明确写下 \"Aim: release + target\"，理由是其近似模型判断反弹接球不可达。视频 2.92–3.32 s 也显示机器人停留原处，右手停在身体侧前方低处，未向墙前下落的反弹球移动，未形成反弹拦截。log step 47 再次以其估计的臂速上限排除后续方案；这支持规划只覆盖前半段，不能证明任务本身无合法解。",
+    "source_zh": "日志"
+   }
+  ],
+  "gap_zh": "没有逐帧状态日志；25 fps 场地侧视图中球和手较小，无法核验毫秒级接触、掌侧 0.12 米几何或精确首次撞墙偏差。首次圆靶命中结合视频约 2.88 s 的球位和 log step 42 本次服务的阶段结果确认；球触地由本次服务终局确认，不把最后一帧当作精确触地时刻。评分字段未报告 DAG 的四项原生指标，均保留 null；agent 在 log step 47 估计的球速和机器人能力上限未获独立验证，不能据此认定所有方案不可行。 log step 46 的 agent 笔记估计反弹落点距机器人约 2–3 米；此距离未经独立测量，不作为视频观察或不可达性的证明。"
+ },
+ "summary_zh": "本次评分回合的球从右手投出并命中墙上圆靶，但视频 2.92–3.32 s 显示反弹球在墙前下落、未到达机器人手部，服务最终因球触地判负，完整“投墙接稳球”节点最终为 0/1。脚本和执行前笔记表明 agent 将策略限于释放与命中，没有规划反弹接球的衔接。缺少状态日志和原生速度、偏差指标，不能验证其关于机器人无法完成任务的判断。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 2,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "fix",
+    "conf": 7,
+    "summary_zh": "节点未完成、断点 wall_throw_caught、没建立接触与推理/规划的大方向与视频、服务读数一致；但证据中“反弹球在机器人伸出的手前方下落”与帧不符：球在墙前靠近圆靶处落下，离机器人 2–3 m，手从未与球接近。",
+    "n_issues": 1,
+    "n_major": 1
+   },
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 2,
+    "verdict": "agree",
+    "conf": 7,
+    "summary_zh": "节点未完成、断点 wall_throw_caught、没建立接触与推理/规划均与视频和 log step 42 的服务读数一致；上一轮指出的球手位置表述已改正，未发现新的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T01:01:26Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--qdgxvtr/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--qdgxvtr/04_review_r1"
+   },
+   {
+    "stage": "reconcile",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--qdgxvtr/05_reconcile_r1"
+   },
+   {
+    "stage": "traj_review",
+    "round": 2,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--qdgxvtr/04_review_r2"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "4f2bf25c53392a45e4c7f34a79092f81ec941cee4e7cdf7f3ca9f2b5c049eea7",
+   "video_sha256": "d1740bed72b158f4a05fbd65174f7ee30e119dbd559bc8e7c990eb4f973b2ec7",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}
diff --git a/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--yqqnnh5.json b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--yqqnnh5.json
new file mode 100644
index 0000000000000000000000000000000000000000..4f570bd7c4259b319d582854d22f1ed2d139fec9
--- /dev/null
+++ b/phase2/embodied-olympics-fun/wall_ball/trajectories/embodied-olympics-fun-wall-ball-i00-standard--yqqnnh5.json
@@ -0,0 +1,215 @@
+{
+ "schema": "efe-trajectory-labels/1",
+ "taxonomy": "v2",
+ "runId": "embodied-olympics-fun-wall-ball-i00-standard--yqqnnh5",
+ "taskKey": "embodied-olympics-fun/wall_ball",
+ "dag_sha256": "6dcdbb6a90a80c19278167c814f0c4df804dc170ce766f3786fd7c7e3c804638",
+ "outcome": {
+  "success": 0,
+  "reward": 0.0,
+  "source": "catalog",
+  "raw": {
+   "success": 0,
+   "live_success": 0,
+   "graded_by_live": 1,
+   "replays": 0,
+   "live_rule": 1,
+   "live_verdict_error": 0,
+   "finalized": 1,
+   "missing_log": 0,
+   "infra_error": 0,
+   "grader_error": 0,
+   "sim_error": 0,
+   "sim_crashed_on_call": 0,
+   "instance_mismatch": 0,
+   "n_actions": 131,
+   "sim_steps": 698,
+   "calls": 13,
+   "rows": 698,
+   "resets": 5,
+   "reset_budget": 5,
+   "sim_startups": 1,
+   "elapsed_s": 3760.6,
+   "video_rendered": 1,
+   "first_success": 0,
+   "first_reason_code": 2,
+   "resource_enforced": 0,
+   "limits_violation_observed": 0,
+   "resource_monitor_missing": 0,
+   "peak_gpu_memory_gb": 0.0,
+   "peak_memory_gb": 1.859,
+   "peak_processes": 13,
+   "peak_disk_gb": 0.582,
+   "resource_emergency_kills": 0,
+   "resource_oom_kills": 0,
+   "resource_host_oom_kills": 0,
+   "resource_service_oom_kills": 0
+  }
+ },
+ "video": {
+  "kind": "hf",
+  "url": "https://huggingface.co/datasets/MLL-Lab/embodied-exam-media/resolve/main/runs/embodied-olympics-fun-wall-ball-i00-standard--yqqnnh5/replay.mp4",
+  "duration_s": 1.32,
+  "fps": 25.0,
+  "frames": 33,
+  "sha256": "8ea8b655be1c80127c33aaa30e564eab077adef56ac8dd7cf726ba58ffe94202"
+ },
+ "state": null,
+ "progress": {
+  "method": "visual",
+  "denominator": 1,
+  "base": 0,
+  "peak": 0,
+  "final": 0,
+  "nodes": {
+   "wall_throw_caught": {
+    "method": "visual",
+    "intervals": [],
+    "final": false,
+    "evidence": [
+     {
+      "t": 0.8,
+      "log_step": null,
+      "text_zh": "观察：视频 0.00–0.84 s，机器人双脚站立，右手携黄绿色球随右臂抬起、摆动；视频 0.80 s 球仍贴着右手，尚未完成投墙接球事件。"
+     },
+     {
+      "t": 0.92,
+      "log_step": null,
+      "text_zh": "观察：逐帧检查视频 0.88–1.28 s，球在视频约 0.92 s 与右手分离，沿低于圆靶的路线飞行并下落；未出现命中红白圆靶、反向飞回及被手接稳的过程。"
+     },
+     {
+      "t": 1.28,
+      "log_step": 168,
+      "text_zh": "观察：log step 168 最后一次提交的服务结果为 episode=6、resets_left=0、done=True、outcome=\"the ball touched the ground\"、stages_reached=1/4；与视频 1.28 s 末帧球已离手并降至场地低处一致。完整投墙接稳球节点未达到。"
+     }
+    ],
+    "contact_actual": [
+     {
+      "who": "手指/夹爪",
+      "how": "抓握",
+      "note_zh": "观察：视频 0.72–0.84 s，右手手指围住黄绿色球并随臂摆动；随后球离手。左手虽做预备动作，但未见左手接触球，未形成双手共同持球或反弹接球。"
+     },
+     {
+      "who": "腿/躯干",
+      "how": "支撑",
+      "note_zh": "观察：视频 0.00–1.28 s，机器人双脚支撑地面，站立配合右臂挥投。"
+     }
+    ]
+   }
+  },
+  "constraints": {}
+ },
+ "native_metrics": [
+  {
+   "name": "target_offset_m",
+   "unit": "m",
+   "threshold": 0.25,
+   "direction": "lower_better",
+   "value": null
+  },
+  {
+   "name": "release_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "wall_speed_m_s",
+   "unit": "m/s",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  },
+  {
+   "name": "catch_height_m",
+   "unit": "m",
+   "threshold": null,
+   "direction": null,
+   "value": null
+  }
+ ],
+ "failure": {
+  "breakpoint": {
+   "ref": "wall_throw_caught",
+   "t": 0.92,
+   "log_step": 168,
+   "where": "操作中",
+   "evidence_zh": "观察：视频约 0.92 s，机器人挥动右臂后球已离手；视频 0.96–1.28 s 球在圆靶下方飞行并下落，没有形成命中与回弹。log step 168 的最终 episode=6 服务结果记录球触地、仅达 release 所对应的 1/4 阶段，投墙接稳球步骤未修复即结束。"
+  },
+  "unmet": {
+   "wall_throw_caught": false
+  },
+  "phenomena": [
+   {
+    "cat": "运动",
+    "item": "位姿不准",
+    "t": 0.92,
+    "log_step": 168,
+    "evidence_zh": "观察：视频 0.88–1.28 s，机器人投出的球未飞抵墙上圆靶，而是从手前方低飞、持续下落；机器人随后抬臂，未再持球。log step 168 的服务裁判结果为球触地、stages_reached=1/4；失败发生在向固定圆靶投球的阶段。"
+   }
+  ],
+  "root_causes": [
+   {
+    "cat": "控制",
+    "item": "控制",
+    "evidence_zh": "推断：log step 167–168 中，agent 将前两种投掷姿态的经验混合结果写成固定关节时间序列，设置右手开启时刻为计划仿真 0.86 s，并一次提交执行；视频 0.88–1.28 s 显示实际出球轨迹低于圆靶并下落，服务在 log step 168 返回球触地。目标是命中固定圆靶，但该挥臂与释放组合未实现所需的出球运动，因此归因于控制；现有证据不足以细分为握持几何、轨迹跟踪或释放参数问题。",
+    "source_zh": "日志 / 视频"
+   }
+  ],
+  "gap_zh": "没有状态日志或逐帧裁判/接触记录，25 fps 场地画面不能精确核实最后接触、0.05 秒释放窗口及具体出球速度。log step 169 的所谓 ball 位置与速度由 agent 自建手部正运动学计算，球离手后不能当作球的实测轨迹；因此不采用 agent 在 log step 179 自述的出球速度、角度或“不可能开环成功”作为已证事实。四项原生诊断指标均未在 reward 中报告，故填 null。"
+ },
+ "summary_zh": "计分视频对应 log step 168 最后一次重置后的第六次尝试：机器人右手握球挥投并释放，但未命中圆靶，也未完成反弹接稳，节点终态为 0/1。断点在投墙接稳球的操作中，视频约 0.92 s 后球低飞下落，服务以球触地结束；现象为位姿不准，固定挥臂与释放组合未产生所需出球轨迹，根因保留为控制。缺少逐帧接触与裁判状态，无法可靠细分释放时机、握球接触和轨迹跟踪的影响。",
+ "conf": 8,
+ "review": {
+  "status": "agreed",
+  "rounds": 1,
+  "reviewers": [
+   {
+    "engine": "claude",
+    "model": "claude-sonnet-5-5",
+    "round": 1,
+    "verdict": "agree",
+    "conf": 6,
+    "summary_zh": "视频仅 1.32 s，球约 0.92 s 离手后低飞落地，节点未达、断点与 log step 168 服务结果一致；现象（位姿不准）与根因（控制，标为推断并说明无法细分）有证据支撑，未发现需改的错误。",
+    "n_issues": 0,
+    "n_major": 0
+   }
+  ],
+  "open_zh": []
+ },
+ "provenance": {
+  "prompt": "prompts/03_trajectory.md",
+  "prompt_sha256": "2385b35930667330fc4ea181a15ea0818b49cfeae0df1700b8b08c988e72ad8e",
+  "engine": "codex",
+  "model": "gpt-6.1-sol",
+  "effort": "high",
+  "created": "2026-10-31T00:49:26Z",
+  "review_prompt": "prompts/04_trajectory_review.md",
+  "review_prompt_sha256": "9afbc4ac19b1e1cf03732c8961ec46a618d682ee229343f66e3316794b14f9ba",
+  "reconcile_prompt": "prompts/05_reconcile.md",
+  "reconcile_prompt_sha256": "8c827af3ae3f77d09229399fb41283c50a8f725302cbe3b263b8051213dde2b2",
+  "calls": [
+   {
+    "stage": "traj",
+    "round": 0,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--yqqnnh5/03_label"
+   },
+   {
+    "stage": "traj_review",
+    "round": 1,
+    "ok": true,
+    "space": "work/runs/embodied-olympics-fun-wall-ball-i00-standard--yqqnnh5/04_review_r1"
+   }
+  ],
+  "review_engine": "claude",
+  "review_model": "claude-sonnet-5-5",
+  "inputs": {
+   "run_json_sha256": "f8b905c449747f36263b6ee019b84f42a71c6bd1722e5196202ce866f03c2129",
+   "video_sha256": "8ea8b655be1c80127c33aaa30e564eab077adef56ac8dd7cf726ba58ffe94202",
+   "states_sha256": null,
+   "video_choice": "HF replay.mp4"
+  }
+ }
+}