example_id stringlengths 16 123 | messages listlengths 14 3.37k | metadata dict | target stringlengths 2 27.4k |
|---|---|---|---|
hc2-api-near-twin-h10-r235-n0541-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring... | FileNotFoundError: configuration audit {'write_files': ['notifier-replicas.conf', 'billing-cutover.conf', 'billing-warmup.conf', 'ingest-replicas.conf', 'notifier-replicas.conf', 'billing-warmup.conf', 'billing-cutover.conf', 'ingest-replicas.conf', 'billing-cutover.conf', 'notifier-replicas.conf', 'ingest-replicas.con... |
hc3-api-replay-h3-r500-n0008-GorillaFileSystem-2-replay_determinism_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "replay_determinism_probe",
"primary_sub_eval": "replay_determinism",
"radius_level": "L0",
"s... | {"files": [{"file_name": "scheduler-cutover.conf", "canonical_content": "quartz-fjord-64"}, {"file_name": "scheduler-limits.conf", "canonical_content": "juniper-harbor-42"}, {"file_name": "ingest-routing.conf", "canonical_content": "harbor-vector-19"}, {"file_name": "registry-shards.conf", "canonical_content": "harbor-... |
code-format_stability_probe-h10-r1-format_stability_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "format_stability_probe",
"primary_sub_eval": "format_stability",
"radius_level": "L0",
"scoring":... | [index-numstat]
[worktree-numstat]
4 2 telemetry/fee_rounding.py
1 1 telemetry/retry_backoff.py
3 1 telemetry/session_window.py
1 1 telemetry/worker_pool.py
[selected-index]
[selected-worktree]
diff --git a/telemetry/session_window.py b/telemetry/session_window.py
--- a/telemetry/session_window.py
+++ b/telemetry/sessi... |
hc3-db-computation-h10-r511-n0045-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:04ede112e26cc0c601bbd6dfc26e4e15c77d6b8bbff6e271a7078b59069b4311",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring": ... | [{"level_fold": 92, "n": 3, "total_stock": 64}] |
pb1-game-computation-h10-r4000-n0001-3-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 181. |
pb2-game-computation-h10-r12015-n0125-3-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 21,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 121. |
ar2g-game-quota-recovery-h10-r40004-n0017-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring... | banana is full; crafting quota exceeded. |
ar2g-game-quota-recovery-h10-r40014-n0057-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring... | banana is full; crafting quota exceeded. |
hc2-api-near-twin-h10-r172-n0289-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring... | FileNotFoundError: configuration audit {'write_files': ['search-quotas.conf', 'registry-routing.conf', 'scheduler-retention.conf', 'indexer-routing.conf', 'search-quotas.conf', 'scheduler-retention.conf', 'registry-routing.conf', 'indexer-routing.conf', 'registry-routing.conf', 'search-quotas.conf', 'indexer-routing.co... |
game-computation_probe-h10-r0-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 110. |
pb1-game-temporal-h3-r4003-n0040-4-rotate_first-temporal_ordering_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "temporal_ordering_probe",
"primary_sub_eval": "temporal_ordering",
"radius_level": "L2",
"scoring"... | You open the refrigerator, revealing a butterfly, an iron, a broom and a Henderson's style key. |
pb1-game-computation-h10-r4006-n0049-3-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 113. |
pb2-game-temporal-h3-r12008-n0100-4-rotate_first-temporal_ordering_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "temporal_ordering_probe",
"primary_sub_eval": "temporal_ordering",
"radius_level": "L2",
"scoring"... | You open the portmanteau, revealing a worm, an insect, a type L key, a blackberry and a cabbage. |
pb1-game-computation-h10-r4007-n0057-3-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 20,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 152. |
pb1-game-computation-h10-r4004-n0037-3-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 216. |
hc3-api-replay-h3-r502-n0032-GorillaFileSystem-2-replay_determinism_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "replay_determinism_probe",
"primary_sub_eval": "replay_determinism",
"radius_level": "L0",
"s... | {"files": [{"file_name": "scheduler-shards.conf", "canonical_content": "indigo-fjord-40"}, {"file_name": "search-routing.conf", "canonical_content": "delta-nimbus-40"}, {"file_name": "registry-backpressure.conf", "canonical_content": "delta-fjord-55"}, {"file_name": "router-quotas.conf", "canonical_content": "nimbus-ba... |
api-replay_determinism_probe-h3-r0-GorillaFileSystem-2-replay_determinism_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:ef410ddd07d6258c8fbb441b16af1bfd87ef6bc3c60db49590ff764e2bc84ed8",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "replay_determinism_probe",
"primary_sub_eval": "replay_determinism",
"radius_level": "L0",
"s... | {"files": [{"file_name": "billing-quotas.conf", "canonical_content": "umber-nimbus-63"}, {"file_name": "registry-replicas.conf", "canonical_content": "amber-juniper-53"}, {"file_name": "gateway-limits.conf", "canonical_content": "juniper-onyx-03"}, {"file_name": "billing-cutover.conf", "canonical_content": "onyx-harbor... |
ar1-api-replay-h3-r30007-n0092-GorillaFileSystem-2-replay_determinism_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "replay_determinism_probe",
"primary_sub_eval": "replay_determinism",
"radius_level": "L0",
"s... | {"files": [{"file_name": "router-routing.conf", "canonical_content": "amber-onyx-75"}, {"file_name": "ingest-quotas.conf", "canonical_content": "amber-amber-02"}, {"file_name": "ledger-warmup.conf", "canonical_content": "onyx-juniper-85"}, {"file_name": "scheduler-retention.conf", "canonical_content": "zephyr-fjord-08"... |
ar1-api-replay-h3-r30008-n0104-GorillaFileSystem-2-replay_determinism_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "replay_determinism_probe",
"primary_sub_eval": "replay_determinism",
"radius_level": "L0",
"s... | {"files": [{"file_name": "ledger-quotas.conf", "canonical_content": "amber-quartz-06"}, {"file_name": "scheduler-replicas.conf", "canonical_content": "basalt-harbor-80"}, {"file_name": "billing-limits.conf", "canonical_content": "sable-harbor-20"}, {"file_name": "registry-cutover.conf", "canonical_content": "fjord-sabl... |
pb1-game-temporal-h3-r4000-n0004-4-rotate_first-temporal_ordering_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "temporal_ordering_probe",
"primary_sub_eval": "temporal_ordering",
"radius_level": "L2",
"scoring"... | You open the Henderson's style box, revealing a fork, a butterfly, an Advent Calendar and a salad. |
pb1-game-computation-h10-r4000-n0005-3-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 190. |
math-inverse_pair-h30-r0-inverse_pair-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "math_and_science",
"domain_legacy": "math",
"executor_image": "sha256:d28e201a98758cc35fe4a0300520f92192a3d3476f9703c72460935508cc2e3f",
"h_reachback": 31,
"item_shape": "output_prediction",
"operator": "inverse_pair",
"primary_sub_eval": "irreversibility",
"radius_level": "L0",
"scoring"... | ['wm_group_audit', 71*(x - 9)*(x + 3) + 83*(x - 9)*(x + 5) + 31*(x - 7)*(x + 2) + 59*(x - 6)*(x + 6) + 17*(x - 6)*(x + 7) + 43*(x - 5)*(x + 3) + 109*(x - 4)*(x + 3) + 127*(x - 4)*(x + 7) + 149*(x - 4)*(x + 8) + 97*(x - 2)*(x + 4), (x - 9)**2*(x + 3)**2 - (x - 9)**2*(x + 5)**2 - (x - 7)**2*(x + 2)**2 - (x - 6)**2*(x + 6... |
hc3-api-replay-h3-r503-n0044-GorillaFileSystem-2-replay_determinism_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "replay_determinism_probe",
"primary_sub_eval": "replay_determinism",
"radius_level": "L0",
"s... | {"files": [{"file_name": "ingest-backpressure.conf", "canonical_content": "krypton-umber-50"}, {"file_name": "registry-routing.conf", "canonical_content": "pearl-garnet-99"}, {"file_name": "billing-retention.conf", "canonical_content": "juniper-basalt-85"}, {"file_name": "ingest-cutover.conf", "canonical_content": "jun... |
hc3-api-replay-h3-r504-n0056-GorillaFileSystem-2-replay_determinism_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "replay_determinism_probe",
"primary_sub_eval": "replay_determinism",
"radius_level": "L0",
"s... | {"files": [{"file_name": "ingest-retention.conf", "canonical_content": "willow-basalt-66"}, {"file_name": "scheduler-routing.conf", "canonical_content": "harbor-krypton-23"}, {"file_name": "router-shards.conf", "canonical_content": "pearl-fjord-46"}, {"file_name": "ingest-shards.conf", "canonical_content": "umber-onyx-... |
pb1-game-computation-h10-r4008-n0065-3-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 83. |
pb2-api-error-h10-r12002-n0029-GorillaFileSystem-2-error_fidelity_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "error_fidelity_probe",
"primary_sub_eval": "error_fidelity",
"radius_level": "L0",
"scoring"... | FileNotFoundError: wc_many completed=[{"file_name":"ledger-quotas.conf","lines":2,"bytes":33},{"file_name":"router-timeouts.conf","lines":1,"bytes":16},{"file_name":"ledger-warmup.conf","lines":3,"bytes":50}]; wc: billing-backpressure.conf: No such file or directory |
ar1-api-replay-h3-r30001-n0020-GorillaFileSystem-2-replay_determinism_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "replay_determinism_probe",
"primary_sub_eval": "replay_determinism",
"radius_level": "L0",
"s... | {"files": [{"file_name": "billing-quotas.conf", "canonical_content": "onyx-russet-18"}, {"file_name": "notifier-quotas.conf", "canonical_content": "basalt-ember-66"}, {"file_name": "search-retention.conf", "canonical_content": "russet-quartz-61"}, {"file_name": "scheduler-retention.conf", "canonical_content": "juniper-... |
code-transition_correctness_probe-h100-r1-transition_correctness_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "transition_correctness_probe",
"primary_sub_eval": "transition_correctness",
"radius_level": "L0",... | M dispatch/worker_pool.py
|
hc3-game-computation-h10-r508-n0065-3-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 92. |
pb2-game-computation-h10-r12006-n0053-3-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 90. |
hc3-api-replay-h3-r507-n0092-GorillaFileSystem-2-replay_determinism_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "replay_determinism_probe",
"primary_sub_eval": "replay_determinism",
"radius_level": "L0",
"s... | {"files": [{"file_name": "notifier-retention.conf", "canonical_content": "quartz-fjord-37"}, {"file_name": "gateway-cutover.conf", "canonical_content": "willow-sable-13"}, {"file_name": "registry-timeouts.conf", "canonical_content": "fjord-cedar-88"}, {"file_name": "billing-limits.conf", "canonical_content": "umber-kry... |
ar1-api-replay-h3-r30010-n0128-GorillaFileSystem-2-replay_determinism_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "replay_determinism_probe",
"primary_sub_eval": "replay_determinism",
"radius_level": "L0",
"s... | {"files": [{"file_name": "scheduler-limits.conf", "canonical_content": "lumen-willow-47"}, {"file_name": "router-quotas.conf", "canonical_content": "delta-willow-41"}, {"file_name": "search-replicas.conf", "canonical_content": "sable-onyx-00"}, {"file_name": "ledger-limits.conf", "canonical_content": "zephyr-onyx-62"}]... |
hc3-db-inverse-h3-r509-n0036-inverse_pair-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:04ede112e26cc0c601bbd6dfc26e4e15c77d6b8bbff6e271a7078b59069b4311",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "inverse_pair",
"primary_sub_eval": "irreversibility",
"radius_level": "L0",
"scoring": {
"adjudi... | [{"focus_id": 100062, "group_fold": 39, "moves": 0, "order_id": 100061, "pair_code": 43, "stage": 83, "status": "held"}, {"focus_id": 100062, "group_fold": 39, "moves": 1, "order_id": 100062, "pair_code": 43, "stage": 70, "status": "pending"}, {"focus_id": 100062, "group_fold": 39, "moves": 0, "order_id": 100063, "pair... |
pb1-game-computation-h10-r4004-n0033-3-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 149. |
pb1-game-temporal-h3-r4000-n0000-3-stack-temporal_ordering_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "temporal_ordering_probe",
"primary_sub_eval": "temporal_ordering",
"radius_level": "L2",
"scoring"... | You open the cabinet, revealing a mop, a sock and a mug. |
hc2-api-near-twin-h3-r181-n0324-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['search-backpressure.conf', 'search-limits.conf', 'indexer-limits.conf', 'notifier-shards.conf', 'search-backpressure.conf', 'indexer-limits.conf', 'search-limits.conf', 'notifier-shards.conf', 'search-limits.conf', 'search-backpressure.conf', 'notifier-shards.con... |
pb2-game-temporal-h3-r12009-n0108-3-stack-temporal_ordering_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "temporal_ordering_probe",
"primary_sub_eval": "temporal_ordering",
"radius_level": "L2",
"scoring"... | You open the case, revealing a type I passkey, a glove and an iron. |
pb2-game-computation-h3-r12007-n0056-2-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 18,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 157. |
hc1-api-near-twin-h3-r36-n0124-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['billing-routing.conf', 'ingest-shards.conf', 'billing-cutover.conf', 'router-warmup.conf', 'billing-routing.conf', 'billing-cutover.conf', 'ingest-shards.conf', 'router-warmup.conf', 'ingest-shards.conf', 'billing-routing.conf', 'router-warmup.conf', 'billing-cut... |
ssp-simpy_activation_cancellation_counterfactual-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "enterprise",
"executor_image": "sha256:5d8647ffedd965e626f74bea080da4d5e6c82c14364220b67d08fc8626cd90cb",
"h_reachback": 59,
"item_shape": "output_prediction",
"operator": "simpy_activation_cancellation_counterfactual",
"primary_sub_eval": "write_read_persistenc... | {"completed_count":20,"missing_count":10,"omit_sequence":39,"queried_makespan":75,"records":[{"completion":10,"job":"job-00","preemptions":0,"stages":[{"finished_at":6,"resource":"reactor","stage":0},{"finished_at":10,"resource":"separator","stage":1}],"status":"completed","total_wait":0},{"completion":null,"job":"job-... |
ar1-shell-entity-delete-h10-r30015-n0061-entity_delete_then_use-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "shell",
"executor_image": "alpine@sha256:6baf43584bcb78f2e5847d1de515f23499913ac9f12bdf834811a3145eb11ca1",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "entity_delete_then_use",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
... | audit_root='/work/etc'
slot1_key=collectd/pool.conf
slot1_state=present
slot1_revision=67
slot2_key=haproxy/workers.conf
slot2_state=present
slot2_revision=47
slot3_key=keepalived/upstream.conf
slot3_state=present
slot3_revision=40
slot4_key=memcached/cache.conf
slot4_state=present
slot4_revision=51
live_total=5
audit_... |
hc1-api-near-twin-h3-r23-n0072-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['indexer-routing.conf', 'ledger-routing.conf', 'gateway-replicas.conf', 'registry-warmup.conf', 'indexer-routing.conf', 'gateway-replicas.conf', 'ledger-routing.conf', 'registry-warmup.conf', 'ledger-routing.conf', 'indexer-routing.conf', 'registry-warmup.conf', '... |
hc1-api-near-twin-h3-r11-n0024-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['ingest-retention.conf', 'billing-routing.conf', 'ingest-quotas.conf', 'ingest-limits.conf', 'ingest-retention.conf', 'ingest-quotas.conf', 'billing-routing.conf', 'ingest-limits.conf', 'billing-routing.conf', 'ingest-retention.conf', 'ingest-limits.conf', 'ingest... |
game-quota_enforcement_probe-h10-r0-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:723f8e8dae1d98deb7cc38d7ef97ffedcfdf48e2d2260744422abda41a3639d9",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring... | sandwich is full; crafting quota exceeded. |
hc2-api-near-twin-h3-r255-n0620-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['gateway-cutover.conf', 'notifier-quotas.conf', 'indexer-cutover.conf', 'ingest-limits.conf', 'gateway-cutover.conf', 'indexer-cutover.conf', 'notifier-quotas.conf', 'ingest-limits.conf', 'notifier-quotas.conf', 'gateway-cutover.conf', 'ingest-limits.conf', 'index... |
db-radius-L0-r0-radius-L0-precision-2 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:e2e30ba2e49e1d4740e29b623fa1446c4f61ee0698b1300fc5f009ffbfd47a29",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L0",
"scoring":... | [{"id": 100026, "selected": 0, "shipped_count": 0, "status": "harbor-russet-65", "status_path": "harbor-russet-65>garnet-harbor-89>onyx-isthmus-91", "total_cents": 0, "weighted_state": 986}, {"id": 100027, "selected": 1, "shipped_count": 0, "status": "garnet-harbor-89", "status_path": "harbor-russet-65>garnet-harbor-89... |
api-write_then_read-h10-r1-GorillaFileSystem-0-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 2,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scorin... | {"audit": {"write_files": ["notifier-cutover.conf", "registry-shards.conf", "registry-timeouts.conf", "billing-replicas.conf", "registry-shards.conf", "registry-timeouts.conf", "billing-replicas.conf", "notifier-cutover.conf", "notifier-cutover.conf", "registry-timeouts.conf", "notifier-cutover.conf", "billing-replicas... |
hc3-code-format-h3-r505-n0020-format_stability_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "format_stability_probe",
"primary_sub_eval": "format_stability",
"radius_level": "L0",
"scoring": ... | [index-numstat]
1 1 catalog/payload_limit.py
3 1 catalog/shard_map.py
3 1 catalog/worker_pool.py
[worktree-numstat]
4 2 catalog/digest_window.py
3 1 catalog/payload_limit.py
1 1 catalog/shard_map.py
1 1 catalog/worker_pool.py
[selected-index]
diff --git a/catalog/payload_limit.py b/catalog/payload_limit.py
--- a/catalo... |
hc3-game-computation-h10-r506-n0049-3-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 84. |
hc2-api-near-twin-h3-r111-n0044-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['ingest-quotas.conf', 'search-shards.conf', 'gateway-shards.conf', 'notifier-limits.conf', 'ingest-quotas.conf', 'gateway-shards.conf', 'search-shards.conf', 'notifier-limits.conf', 'search-shards.conf', 'ingest-quotas.conf', 'notifier-limits.conf', 'gateway-shard... |
hc1-api-near-twin-h3-r10-n0020-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['ledger-retention.conf', 'gateway-warmup.conf', 'billing-routing.conf', 'ingest-timeouts.conf', 'ledger-retention.conf', 'billing-routing.conf', 'gateway-warmup.conf', 'ingest-timeouts.conf', 'gateway-warmup.conf', 'ledger-retention.conf', 'ingest-timeouts.conf', ... |
hc2-api-near-twin-h3-r137-n0148-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['scheduler-warmup.conf', 'search-replicas.conf', 'indexer-warmup.conf', 'router-shards.conf', 'scheduler-warmup.conf', 'indexer-warmup.conf', 'search-replicas.conf', 'router-shards.conf', 'search-replicas.conf', 'scheduler-warmup.conf', 'router-shards.conf', 'inde... |
hc3-game-computation-h3-r501-n0008-2-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 16,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 59. |
hc3-code-format-h3-r504-n0016-format_stability_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "format_stability_probe",
"primary_sub_eval": "format_stability",
"radius_level": "L0",
"scoring": ... | [index-numstat]
3 1 dispatch/digest_window.py
3 1 dispatch/fee_rounding.py
1 1 dispatch/worker_pool.py
[worktree-numstat]
1 1 dispatch/digest_window.py
1 1 dispatch/fee_rounding.py
4 2 dispatch/shard_map.py
3 1 dispatch/worker_pool.py
[selected-index]
diff --git a/dispatch/worker_pool.py b/dispatch/worker_pool.py
--- a... |
hc3-code-format-h3-r500-n0000-format_stability_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "format_stability_probe",
"primary_sub_eval": "format_stability",
"radius_level": "L0",
"scoring": ... | [index-numstat]
3 1 catalog/digest_window.py
3 1 catalog/payload_limit.py
1 1 catalog/worker_pool.py
[worktree-numstat]
1 1 catalog/digest_window.py
1 1 catalog/payload_limit.py
4 2 catalog/queue_drain.py
3 1 catalog/worker_pool.py
[selected-index]
diff --git a/catalog/worker_pool.py b/catalog/worker_pool.py
--- a/cata... |
hc1-api-near-twin-h3-r20-n0060-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['billing-shards.conf', 'registry-replicas.conf', 'scheduler-replicas.conf', 'billing-limits.conf', 'billing-shards.conf', 'scheduler-replicas.conf', 'registry-replicas.conf', 'billing-limits.conf', 'registry-replicas.conf', 'billing-shards.conf', 'billing-limits.c... |
hc2-api-near-twin-h3-r248-n0592-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['notifier-warmup.conf', 'ingest-replicas.conf', 'gateway-cutover.conf', 'notifier-replicas.conf', 'notifier-warmup.conf', 'gateway-cutover.conf', 'ingest-replicas.conf', 'notifier-replicas.conf', 'ingest-replicas.conf', 'notifier-warmup.conf', 'notifier-replicas.c... |
hc2-api-near-twin-h3-r259-n0636-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['search-backpressure.conf', 'indexer-cutover.conf', 'notifier-replicas.conf', 'billing-warmup.conf', 'search-backpressure.conf', 'notifier-replicas.conf', 'indexer-cutover.conf', 'billing-warmup.conf', 'indexer-cutover.conf', 'search-backpressure.conf', 'billing-w... |
hc2-api-near-twin-h3-r144-n0176-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['search-quotas.conf', 'notifier-routing.conf', 'registry-shards.conf', 'router-cutover.conf', 'search-quotas.conf', 'registry-shards.conf', 'notifier-routing.conf', 'router-cutover.conf', 'notifier-routing.conf', 'search-quotas.conf', 'router-cutover.conf', 'regis... |
api-access_control_probe-h10-r0-TicketAPI-1-access_control_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "access_control_probe",
"primary_sub_eval": "access_control",
"radius_level": "L0",
"scoring"... | [{"error": "User not authenticated. Please log in to view tickets."}] |
api-access_control_probe-h100-r0-TicketAPI-1-access_control_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "access_control_probe",
"primary_sub_eval": "access_control",
"radius_level": "L0",
"scoring... | [{"error": "User not authenticated. Please log in to view tickets."}] |
hc2-api-near-twin-h3-r177-n0308-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['search-timeouts.conf', 'billing-warmup.conf', 'router-shards.conf', 'scheduler-backpressure.conf', 'search-timeouts.conf', 'router-shards.conf', 'billing-warmup.conf', 'scheduler-backpressure.conf', 'billing-warmup.conf', 'search-timeouts.conf', 'scheduler-backpr... |
math-write_then_read-h30-r0-write_then_read-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "math_and_science",
"domain_legacy": "math",
"executor_image": "sha256:d28e201a98758cc35fe4a0300520f92192a3d3476f9703c72460935508cc2e3f",
"h_reachback": 44,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
... | ['wm_group_audit', 327464/55, 376929/4400, 22941/220, False, 16, 97210, 264168] |
hc1-api-near-twin-h3-r30-n0100-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['router-quotas.conf', 'indexer-warmup.conf', 'router-routing.conf', 'billing-retention.conf', 'router-quotas.conf', 'router-routing.conf', 'indexer-warmup.conf', 'billing-retention.conf', 'indexer-warmup.conf', 'router-quotas.conf', 'billing-retention.conf', 'rout... |
hc1-shell-transition-h10-r6-n0005-transition_correctness_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "shell",
"executor_image": "alpine@sha256:6baf43584bcb78f2e5847d1de515f23499913ac9f12bdf834811a3145eb11ca1",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "transition_correctness_probe",
"primary_sub_eval": "transition_correctness",
"radius_level"... | slot1_service=collectd/cache.conf
slot1_state=standby
slot1_gen=5
slot2_service=keepalived/main.conf
slot2_state=disabled
slot2_gen=7
slot3_service=keepalived/upstream.conf
slot3_state=disabled
slot3_gen=5
slot4_service=logrotate/limits.conf
slot4_state=enabled
slot4_gen=31
audit_events=12
audit_applied=6
audit_rejecte... |
ar1-shell-entity-delete-h10-r30000-n0001-entity_delete_then_use-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "shell",
"executor_image": "alpine@sha256:6baf43584bcb78f2e5847d1de515f23499913ac9f12bdf834811a3145eb11ca1",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "entity_delete_then_use",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
... | audit_root='/work/etc'
slot1_key=collectd/server.conf
slot1_state=present
slot1_revision=20
slot2_key=keepalived/limits.conf
slot2_state=present
slot2_revision=94
slot3_key=memcached/network.conf
slot3_state=present
slot3_revision=6
slot4_key=memcached/pool.conf
slot4_state=present
slot4_revision=41
live_total=5
audit_... |
hc2-api-near-twin-h3-r224-n0496-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['scheduler-retention.conf', 'search-timeouts.conf', 'ledger-limits.conf', 'billing-quotas.conf', 'scheduler-retention.conf', 'ledger-limits.conf', 'search-timeouts.conf', 'billing-quotas.conf', 'search-timeouts.conf', 'scheduler-retention.conf', 'billing-quotas.co... |
ar1-shell-entity-delete-h10-r30006-n0025-entity_delete_then_use-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "shell",
"executor_image": "alpine@sha256:6baf43584bcb78f2e5847d1de515f23499913ac9f12bdf834811a3145eb11ca1",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "entity_delete_then_use",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
... | audit_root='/work/etc'
slot1_key=chrony/auth.conf
slot1_state=present
slot1_revision=2
slot2_key=logrotate/cache.conf
slot2_state=present
slot2_revision=10
slot3_key=memcached/cache.conf
slot3_state=missing
slot3_revision=25
slot4_key=postfix/listeners.conf
slot4_state=present
slot4_revision=78
live_total=5
audit_event... |
hc2-api-near-twin-h3-r193-n0372-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['scheduler-retention.conf', 'billing-quotas.conf', 'scheduler-warmup.conf', 'billing-warmup.conf', 'scheduler-retention.conf', 'scheduler-warmup.conf', 'billing-quotas.conf', 'billing-warmup.conf', 'billing-quotas.conf', 'scheduler-retention.conf', 'billing-warmup... |
hc2-api-near-twin-h3-r234-n0536-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['indexer-shards.conf', 'billing-retention.conf', 'billing-routing.conf', 'router-replicas.conf', 'indexer-shards.conf', 'billing-routing.conf', 'billing-retention.conf', 'router-replicas.conf', 'billing-retention.conf', 'indexer-shards.conf', 'router-replicas.conf... |
ar2g-game-quota-recovery-h10-r40001-n0005-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring... | berry is full; crafting quota exceeded. |
ar1-shell-entity-delete-h10-r30012-n0049-entity_delete_then_use-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "shell",
"executor_image": "alpine@sha256:6baf43584bcb78f2e5847d1de515f23499913ac9f12bdf834811a3145eb11ca1",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "entity_delete_then_use",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
... | audit_root='/work/etc'
slot1_key=postfix/listeners.conf
slot1_state=present
slot1_revision=39
slot2_key=postfix/workers.conf
slot2_state=present
slot2_revision=31
slot3_key=redis/main.conf
slot3_state=present
slot3_revision=25
slot4_key=redis/server.conf
slot4_state=present
slot4_revision=17
live_total=5
audit_events=1... |
ar2g-game-quota-recovery-h10-r40000-n0001-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring... | cauliflower is full; crafting quota exceeded. |
hc3-game-quota-h10-r505-n0021-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring... | cauliflower is full; crafting quota exceeded. |
hc3-code-format-h3-r503-n0012-format_stability_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "format_stability_probe",
"primary_sub_eval": "format_stability",
"radius_level": "L0",
"scoring": ... | [index-numstat]
[worktree-numstat]
1 1 billing/cache_expiry.py
4 2 billing/payload_limit.py
3 1 billing/queue_drain.py
1 1 billing/session_window.py
[selected-index]
[selected-worktree]
diff --git a/billing/queue_drain.py b/billing/queue_drain.py
--- a/billing/queue_drain.py
+++ b/billing/queue_drain.py
@@ -3,3 +3,3 @@... |
game-computation_probe-h3-r1-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 10,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 187. |
hc3-game-computation-h3-r509-n0076-2-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 151. |
hc3-code-format-h3-r502-n0008-format_stability_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "format_stability_probe",
"primary_sub_eval": "format_stability",
"radius_level": "L0",
"scoring": ... | [index-numstat]
[worktree-numstat]
4 2 billing/batch_sizing.py
3 1 billing/queue_drain.py
1 1 billing/rate_limit.py
1 1 billing/retry_backoff.py
[selected-index]
[selected-worktree]
diff --git a/billing/queue_drain.py b/billing/queue_drain.py
--- a/billing/queue_drain.py
+++ b/billing/queue_drain.py
@@ -3,3 +3,3 @@
PR... |
pb2-game-computation-h3-r12013-n0104-2-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 13,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 111. |
hc2-api-near-twin-h3-r232-n0528-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['notifier-replicas.conf', 'gateway-cutover.conf', 'router-warmup.conf', 'scheduler-retention.conf', 'notifier-replicas.conf', 'router-warmup.conf', 'gateway-cutover.conf', 'scheduler-retention.conf', 'gateway-cutover.conf', 'notifier-replicas.conf', 'scheduler-ret... |
hc2-api-near-twin-h3-r110-n0040-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['notifier-cutover.conf', 'scheduler-warmup.conf', 'gateway-routing.conf', 'ingest-limits.conf', 'notifier-cutover.conf', 'gateway-routing.conf', 'scheduler-warmup.conf', 'ingest-limits.conf', 'scheduler-warmup.conf', 'notifier-cutover.conf', 'ingest-limits.conf', ... |
hc3-game-computation-h3-r501-n0012-2-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 13,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | Total score: 51. |
hc3-shell-persistence-h100-r510-n0043-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "shell",
"executor_image": "alpine@sha256:6baf43584bcb78f2e5847d1de515f23499913ac9f12bdf834811a3145eb11ca1",
"h_reachback": 15,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"sc... | wm_group_audit current=291 cross=397 events=16 transitions=8 digest=984
|
hc2-api-near-twin-h3-r154-n0216-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring"... | FileNotFoundError: configuration audit {'write_files': ['router-warmup.conf', 'notifier-cutover.conf', 'gateway-cutover.conf', 'router-routing.conf', 'router-warmup.conf', 'gateway-cutover.conf', 'notifier-cutover.conf', 'router-routing.conf', 'notifier-cutover.conf', 'router-warmup.conf', 'router-routing.conf', 'gatew... |
game-transition_correctness_probe-h3-r0-transition_correctness_probe-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:723f8e8dae1d98deb7cc38d7ef97ffedcfdf48e2d2260744422abda41a3639d9",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "transition_correctness_probe",
"primary_sub_eval": "transition_correctness",
"radius_level": "L0",
... | The coffer looks strong, and impossible to crack. You can see inside it.
In the coffer is a dvd. |
ar1-shell-entity-delete-h10-r30007-n0029-entity_delete_then_use-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "shell",
"executor_image": "alpine@sha256:6baf43584bcb78f2e5847d1de515f23499913ac9f12bdf834811a3145eb11ca1",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "entity_delete_then_use",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
... | audit_root='/work/etc'
slot1_key=collectd/listeners.conf
slot1_state=present
slot1_revision=52
slot2_key=haproxy/limits.conf
slot2_state=missing
slot2_revision=16
slot3_key=haproxy/listeners.conf
slot3_state=present
slot3_revision=33
slot4_key=keepalived/auth.conf
slot4_state=present
slot4_revision=12
live_total=5
audi... |
hc3-game-quota-h10-r500-n0001-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring... | legume is full; crafting quota exceeded. |
ar1-shell-temporal-h10-r30035-n0141-temporal_ordering_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "shell",
"executor_image": "alpine@sha256:6baf43584bcb78f2e5847d1de515f23499913ac9f12bdf834811a3145eb11ca1",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "temporal_ordering_probe",
"primary_sub_eval": "temporal_ordering",
"radius_level": "L0",
... | applied_1=chrony/server.conf:rotate-13
applied_2=vsftpd/upstream.conf:sync-23
applied_3=memcached/server.conf:flush-49
applied_4=vsftpd/upstream.conf:compact-47
applied_5=memcached/auth.conf:gc-83
applied_6=memcached/server.conf:sync-12
audit_applied=6
audit_switches=5
audit_digest=341
audit_pending=4
|
game-entity_delete_then_use-h3-r0-entity_delete_then_use-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:723f8e8dae1d98deb7cc38d7ef97ffedcfdf48e2d2260744422abda41a3639d9",
"h_reachback": 9,
"item_shape": "output_prediction",
"operator": "entity_delete_then_use",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring... | You couldn't pay me to eat that ordinary thing. |
hc3-code-inverse-h10-r500-n0001-inverse_pair-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "inverse_pair",
"primary_sub_eval": "irreversibility",
"radius_level": "L0",
"scoring": {
"adj... | ...F [100%]
=========================== short test summary info ============================
FAILED tests/test_worker_pool.py::test_module_constants - AssertionError: ass...
1 failed, 3 passed in Xs
inverse_group_audit selected=catalog/worker_pool.py m... |
hc3-code-inverse-h10-r501-n0005-inverse_pair-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "inverse_pair",
"primary_sub_eval": "irreversibility",
"radius_level": "L0",
"scoring": {
"adj... | ...F [100%]
=========================== short test summary info ============================
FAILED tests/test_batch_sizing.py::test_module_constants - AssertionError: as...
1 failed, 3 passed in Xs
inverse_group_audit selected=ingest/batch_sizing.py m... |
ar2g-game-quota-recovery-h10-r40009-n0037-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring... | coconut is full; crafting quota exceeded. |
game-computation_probe-h3-r0-2-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:723f8e8dae1d98deb7cc38d7ef97ffedcfdf48e2d2260744422abda41a3639d9",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring":... | Total score: 94. |
hc3-code-computation-h10-r504-n0017-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | 8 billing/digest_window.py
16 billing/rate_limit.py
8 billing/audit_trail.py
32 total
|
shell-error_fidelity_probe-h100-r0-error_fidelity_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "shell",
"executor_image": "alpine@sha256:6baf43584bcb78f2e5847d1de515f23499913ac9f12bdf834811a3145eb11ca1",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "error_fidelity_probe",
"primary_sub_eval": "error_fidelity",
"radius_level": "L0",
"scor... | head: /work/etc/memcached/server.conf: No such file or directory
|
hc3-code-computation-h10-r508-n0033-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | 16 dispatch/session_window.py
12 dispatch/queue_drain.py
16 dispatch/batch_sizing.py
44 total
|
ar1-shell-temporal-h3-r30005-n0020-temporal_ordering_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "shell",
"executor_image": "alpine@sha256:6baf43584bcb78f2e5847d1de515f23499913ac9f12bdf834811a3145eb11ca1",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "temporal_ordering_probe",
"primary_sub_eval": "temporal_ordering",
"radius_level": "L0",
"... | applied_1=chrony/pool.conf:reload-35
applied_2=redis/workers.conf:flush-74
applied_3=postfix/server.conf:reload-43
applied_4=haproxy/upstream.conf:compact-73
applied_5=haproxy/upstream.conf:gc-72
applied_6=chrony/pool.conf:flush-89
applied_7=postfix/server.conf:rotate-11
audit_applied=7
audit_switches=5
audit_digest=11... |
hc3-game-computation-h3-r506-n0052-2-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring":... | Total score: 38. |
game-quota_enforcement_probe-h3-r1-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring"... | durian is full; crafting quota exceeded. |
code-computation_probe-h10-r0-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | 12 billing/session_window.py
16 billing/queue_drain.py
8 billing/worker_pool.py
36 total
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.