diff --git "a/data/tasks.jsonl" "b/data/tasks.jsonl"
new file mode 100644--- /dev/null
+++ "b/data/tasks.jsonl"
@@ -0,0 +1,100 @@
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_type_correct", "offending_key_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "the path from analytics-worker to pg-replica is REFUSED at the transport layer rather than timing out, so this is a network policy applied on day 418 and not a capacity problem; a timeout would look like load and a refusal cannot", "fault_detected": true, "fault_type": "misconfig", "offending_key": "pg-replica", "scope": "analytics-connection-refused", "service": "analytics-worker"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_analysis", "code_license": "Apache-2.0", "curation_rationale": ["aiops_analysis", "expert", "9-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-001-rca-analytics-egress-blocked", "llm_judge": false, "origin": "curated", "reference_tool_calls": 9, "required_tools": ["check_network_path", "get_service", "get_ticket", "list_error_events", "query_metrics", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_rca_analytics_egress_blocked", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-163 — Root cause: analytics connections are refused, not slow\n\nGet to the bottom of analytics-connection-refused: not just which service, but the mechanism and the exact setting responsible. The service's own code is not necessarily where this lives — confirm it against the state of the system rather than inferring it from the symptom alone. Submit your finding under scope 'analytics-connection-refused' with the responsible service, the fault type, the offending key, and your evidence.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_type_correct", "offending_key_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-001-rca-analytics-egress-blocked.py"}, "task_id": "dob100-001-rca-analytics-egress-blocked", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_type_correct", "offending_key_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "the analytics_ro role on pg-replica was dropped during the day-418 credential rotation, but the running analytics-worker pods were never restarted and still present it; the database is healthy and so is the service, and the grant between them is not", "fault_detected": true, "fault_type": "misconfig", "offending_key": "analytics_ro", "scope": "analytics-cannot-read-replica", "service": "analytics-worker"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_analysis", "code_license": "Apache-2.0", "curation_rationale": ["aiops_analysis", "expert", "9-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-002-rca-analytics-revoked-grant", "llm_judge": false, "origin": "curated", "reference_tool_calls": 9, "required_tools": ["get_service", "get_ticket", "list_db_grants", "list_error_events", "list_infra", "query_metrics", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_rca_analytics_revoked_grant", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-160 — Root cause: analytics cannot read the replica since the rotation\n\nGet to the bottom of analytics-cannot-read-replica: not just which service, but the mechanism and the exact setting responsible. The service's own code is not necessarily where this lives — confirm it against the state of the system rather than inferring it from the symptom alone. Submit your finding under scope 'analytics-cannot-read-replica' with the responsible service, the fault type, the offending key, and your evidence.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_type_correct", "offending_key_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-002-rca-analytics-revoked-grant.py"}, "task_id": "dob100-002-rca-analytics-revoked-grant", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_type_correct", "offending_key_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "the checkout deployment declares 64 desired replicas and has 6 ready; the 58 that cannot be admitted are Pending with '4 Insufficient cpu' and no node in the cluster has the capacity, so the shortfall is in the requested replica count rather than in the cluster or the service", "fault_detected": true, "fault_type": "misconfig", "offending_key": "desired_replicas", "scope": "checkout-capacity-shortfall", "service": "checkout"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_analysis", "code_license": "Apache-2.0", "curation_rationale": ["aiops_analysis", "expert", "10-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-003-rca-checkout-unschedulable-replicas", "llm_judge": false, "origin": "curated", "reference_tool_calls": 10, "required_tools": ["get_service", "get_ticket", "k8s_deployments_list", "k8s_nodes_list", "k8s_pods_list", "list_error_events", "query_metrics", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_rca_checkout_unschedulable_replicas", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-129 — Root cause: checkout is running at a fraction of capacity\n\nGet to the bottom of checkout-capacity-shortfall: not just which service, but the mechanism and the exact setting responsible. The service's own code is not necessarily where this lives — confirm it against the state of the system rather than inferring it from the symptom alone. Submit your finding under scope 'checkout-capacity-shortfall' with the responsible service, the fault type, the offending key, and your evidence.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_type_correct", "offending_key_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-003-rca-checkout-unschedulable-replicas.py"}, "task_id": "dob100-003-rca-checkout-unschedulable-replicas", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_type_correct", "offending_key_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "both media-service replicas are scheduled on node-b3, whose kubelet reports DiskPressure at 97% of 200Gi ephemeral storage, and one replica has already been Evicted for ephemeral-storage three times; the pods themselves are Running with normal memory and the service's own code is unchanged", "fault_detected": true, "fault_type": "node_unhealthy", "offending_key": "node-b3", "scope": "media-upload-stalls", "service": "media-service"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_analysis", "code_license": "Apache-2.0", "curation_rationale": ["aiops_analysis", "expert", "10-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-004-rca-media-disk-pressure", "llm_judge": false, "origin": "curated", "reference_tool_calls": 10, "required_tools": ["get_service", "get_ticket", "k8s_events_list", "k8s_nodes_list", "k8s_pods_list", "list_error_events", "query_metrics", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_rca_media_disk_pressure", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-125 — Root cause: media uploads stalling\n\nGet to the bottom of media-upload-stalls: not just which service, but the mechanism and the exact setting responsible. The service's own code is not necessarily where this lives — confirm it against the state of the system rather than inferring it from the symptom alone. Submit your finding under scope 'media-upload-stalls' with the responsible service, the fault type, the offending key, and your evidence.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_type_correct", "offending_key_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-004-rca-media-disk-pressure.py"}, "task_id": "dob100-004-rca-media-disk-pressure", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_type_correct", "offending_key_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "node-c1 is Ready=Unknown with a kernel soft lockup on CPU#3 and stopped posting node status 14 minutes ago; notifications-1b7d-hh88 still reports Running because the kubelet stopped reporting, not the pod, so pod phase alone says the service is healthy while it is unreachable", "fault_detected": true, "fault_type": "node_unhealthy", "offending_key": "node-c1", "scope": "notifications-unreachable", "service": "notifications"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_analysis", "code_license": "Apache-2.0", "curation_rationale": ["aiops_analysis", "expert", "10-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-005-rca-notifications-node-deadlock", "llm_judge": false, "origin": "curated", "reference_tool_calls": 10, "required_tools": ["get_service", "get_ticket", "k8s_events_list", "k8s_nodes_list", "k8s_pods_list", "list_error_events", "query_metrics", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_rca_notifications_node_deadlock", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-127 — Root cause: notifications are not being delivered\n\nGet to the bottom of notifications-unreachable: not just which service, but the mechanism and the exact setting responsible. The service's own code is not necessarily where this lives — confirm it against the state of the system rather than inferring it from the symptom alone. Submit your finding under scope 'notifications-unreachable' with the responsible service, the fault type, the offending key, and your evidence.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_type_correct", "offending_key_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-005-rca-notifications-node-deadlock.py"}, "task_id": "dob100-005-rca-notifications-node-deadlock", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_type_correct", "offending_key_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "search-reindex-8c2a-gg77 has been Pending since day 419 with 214 FailedScheduling events reading 0/4 nodes available: no node matches the pod's accelerator=gpu-a100 selector, and no node in the cluster carries that label; the reindex has therefore never run once and the index is stale", "fault_detected": true, "fault_type": "misconfig", "offending_key": "accelerator=gpu-a100", "scope": "search-stale-index", "service": "search"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_analysis", "code_license": "Apache-2.0", "curation_rationale": ["aiops_analysis", "expert", "10-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-006-rca-search-unscheduled-reindex", "llm_judge": false, "origin": "curated", "reference_tool_calls": 10, "required_tools": ["get_service", "get_ticket", "k8s_events_list", "k8s_nodes_list", "k8s_pods_list", "list_error_events", "query_metrics", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_rca_search_unscheduled_reindex", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-126 — Root cause: search results are stale\n\nGet to the bottom of search-stale-index: not just which service, but the mechanism and the exact setting responsible. The service's own code is not necessarily where this lives — confirm it against the state of the system rather than inferring it from the symptom alone. Submit your finding under scope 'search-stale-index' with the responsible service, the fault type, the offending key, and your evidence.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_type_correct", "offending_key_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-006-rca-search-unscheduled-reindex.py"}, "task_id": "dob100-006-rca-search-unscheduled-reindex", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "detection_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "checkout error_rate_pct is 5.5 against a 1.0 SLO while latency_p99_ms 180 is within its 400 SLO", "fault_detected": true, "fault_type": "feature_flag_regression", "offending_key": "instant_refunds", "scope": "checkout", "service": "checkout"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_detection", "code_license": "Apache-2.0", "curation_rationale": ["aiops_detection", "medium", "6-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-007-detect-checkout-errors", "llm_judge": false, "origin": "curated", "reference_tool_calls": 6, "required_tools": ["get_slo_status", "get_ticket", "list_alerts", "query_metrics", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_detect_checkout_errors", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-104 — Detection: is checkout violating its SLOs?\n\nDetermine whether checkout is currently violating any of its service level objectives, and submit the result under scope 'checkout' with the evidence you relied on. Report honestly either way: a clean bill of health is a valid finding.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "detection_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-007-detect-checkout-errors.py"}, "task_id": "dob100-007-detect-checkout-errors", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "detection_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "inventory error_rate_pct is 4.7 against a 1.0 SLO; alarm 9606 is firing", "fault_detected": true, "fault_type": "resource_exhaustion", "offending_key": "db_pool_size", "scope": "inventory", "service": "inventory"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_detection", "code_license": "Apache-2.0", "curation_rationale": ["aiops_detection", "easy", "6-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "easy", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-008-detect-inventory", "llm_judge": false, "origin": "curated", "reference_tool_calls": 6, "required_tools": ["get_slo_status", "get_ticket", "list_alerts", "query_metrics", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_detect_inventory", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-102 — Detection: is inventory violating its SLOs?\n\nDetermine whether inventory is currently violating any of its service level objectives, and submit the result under scope 'inventory' with the evidence you relied on. Report honestly either way: a clean bill of health is a valid finding.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "detection_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-008-detect-inventory.py"}, "task_id": "dob100-008-detect-inventory", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "detection_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "payments error_rate_pct is 4.2 against a 1.0 SLO; alarm 9601 is firing", "fault_detected": true, "fault_type": "missing_retry", "offending_key": "notifications_retry_max_attempts", "scope": "payments", "service": "payments"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_detection", "code_license": "Apache-2.0", "curation_rationale": ["aiops_detection", "easy", "6-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "easy", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-009-detect-payments", "llm_judge": false, "origin": "curated", "reference_tool_calls": 6, "required_tools": ["get_slo_status", "get_ticket", "list_alerts", "query_metrics", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_detect_payments", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-101 — Detection: is payments violating its SLOs?\n\nDetermine whether payments is currently violating any of its service level objectives, and submit the result under scope 'payments' with the evidence you relied on. Report honestly either way: a clean bill of health is a valid finding.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "detection_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-009-detect-payments.py"}, "task_id": "dob100-009-detect-payments", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "detection_correct", "service_identified", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "the status page post 7001 is marked resolved, but checkout is breaching both of its SLOs right now - error_rate_pct 5.5 against a 1.0 threshold and latency_p99_ms 530 against 400 - and alarms 9603 and 9610 are both firing; the status page lags internal state and is not evidence of health", "fault_detected": true, "fault_type": "unclassified", "offending_key": "", "scope": "status-7001-recurrence", "service": "checkout"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_detection", "code_license": "Apache-2.0", "curation_rationale": ["aiops_detection", "expert", "8-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-010-detect-status-page-recurrence", "llm_judge": false, "origin": "curated", "reference_tool_calls": 8, "required_tools": ["get_slo_status", "get_ticket", "list_alerts", "list_status_page_posts", "query_metrics", "submit_diagnosis", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_detect_status_page_recurrence", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-131 — Customers still cannot check out, and the status page says we fixed it\n\nSupport has had eleven tickets this morning from customers who cannot complete a purchase - they get an error at the final step and give up. Our status page has nothing open: the only related post, 'Degraded checkout performance', was marked resolved on day 412. Support wants to know whether to keep telling customers this is their browser.\n\nWork out which service this is about, determine whether it is currently violating any of its service level objectives, and submit the result under scope 'status-7001-recurrence' — naming the service in the `service` field — with the evidence you relied on. Report honestly either way: a clean bill of health is a valid finding, and so is contradicting a system that says everything is fine.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "detection_correct", "service_identified", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-010-detect-status-page-recurrence.py"}, "task_id": "dob100-010-detect-status-page-recurrence", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "detection_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "storefront-web latency_p99_ms is 220 against a 500 SLO and no alarm is firing", "fault_detected": false, "fault_type": "none", "offending_key": "", "scope": "storefront-web", "service": ""}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_detection", "code_license": "Apache-2.0", "curation_rationale": ["aiops_detection", "medium", "6-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-011-detect-storefront-healthy", "llm_judge": false, "origin": "curated", "reference_tool_calls": 6, "required_tools": ["get_slo_status", "get_ticket", "list_alerts", "query_metrics", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_detect_storefront_healthy", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-103 — Detection: is storefront-web violating its SLOs?\n\nDetermine whether storefront-web is currently violating any of its service level objectives, and submit the result under scope 'storefront-web' with the evidence you relied on. Report honestly either way: a clean bill of health is a valid finding.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "detection_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-011-detect-storefront-healthy.py"}, "task_id": "dob100-011-detect-storefront-healthy", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_confirmed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "the kubelet records 47 OOMKilled events and a CrashLoopBackOff on analytics-worker-7d9f-x2k1 at a 512Mi limit against 511Mi usage; the error tracker shows nothing for this service because the process is killed before its SDK can flush", "fault_detected": true, "fault_type": "resource_exhaustion", "offending_key": "memory_limit_mb", "scope": "9609", "service": "analytics-worker"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_localization", "code_license": "Apache-2.0", "curation_rationale": ["aiops_localization", "expert", "7-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-012-localize-analytics-crashloop", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["get_service", "get_slo_status", "get_ticket", "list_alerts", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_localize_analytics_crashloop", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-116 — Localize alarm 9609 — the error tracker shows nothing\n\nAlarm 9609 is firing. Work out which service is actually responsible — the service the alarm names is not automatically the culprit — and submit your finding under scope '9609' with the evidence you relied on.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_confirmed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-012-localize-analytics-crashloop.py"}, "task_id": "dob100-012-localize-analytics-crashloop", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_confirmed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "analytics-worker error_rate_pct 6.0; the consumer OOMs with unlimited queue prefetch", "fault_detected": true, "fault_type": "unbounded_prefetch", "offending_key": "prefetch_count", "scope": "9609", "service": "analytics-worker"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_localization", "code_license": "Apache-2.0", "curation_rationale": ["aiops_localization", "medium", "7-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-013-localize-analytics-errors", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["get_service", "get_slo_status", "get_ticket", "list_alerts", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_localize_analytics_errors", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-113 — Localize alarm 9609 (analytics-worker errors)\n\nAlarm 9609 is firing. Work out which service is actually responsible — the service the alarm names is not automatically the culprit — and submit your finding under scope '9609' with the evidence you relied on.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_confirmed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-013-localize-analytics-errors.py"}, "task_id": "dob100-013-localize-analytics-errors", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_confirmed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "checkout p99 530ms is spent waiting on payments, which blocks for up to 30s on its notifications call; checkout's own configuration is within policy", "fault_detected": true, "fault_type": "misconfig", "offending_key": "notifications_timeout_ms", "scope": "9610", "service": "payments"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_localization", "code_license": "Apache-2.0", "curation_rationale": ["aiops_localization", "hard", "7-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:shortcut,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-014-localize-checkout-latency", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["get_service", "get_slo_status", "get_ticket", "list_alerts", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_localize_checkout_latency", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-115 — Localize alarm 9610 (checkout latency)\n\nAlarm 9610 is firing. Work out which service is actually responsible — the service the alarm names is not automatically the culprit — and submit your finding under scope '9610' with the evidence you relied on.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_confirmed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-014-localize-checkout-latency.py"}, "task_id": "dob100-014-localize-checkout-latency", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_confirmed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "p99 1030ms began immediately after v5.1.0 was promoted; the pool opens a connection per request", "fault_detected": true, "fault_type": "bad_release", "offending_key": "v5.1.0", "scope": "9604", "service": "api-gateway"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_localization", "code_license": "Apache-2.0", "curation_rationale": ["aiops_localization", "medium", "7-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-015-localize-gateway-latency", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["get_service", "get_slo_status", "get_ticket", "list_alerts", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_localize_gateway_latency", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-111 — Localize alarm 9604 (api-gateway latency)\n\nAlarm 9604 is firing. Work out which service is actually responsible — the service the alarm names is not automatically the culprit — and submit your finding under scope '9604' with the evidence you relied on.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_confirmed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-015-localize-gateway-latency.py"}, "task_id": "dob100-015-localize-gateway-latency", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_confirmed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "media-service p99 800ms with every asset request served from the origin object store", "fault_detected": true, "fault_type": "cdn_bypass", "offending_key": "cdn_enabled", "scope": "9607", "service": "media-service"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "aiops_localization", "code_license": "Apache-2.0", "curation_rationale": ["aiops_localization", "medium", "7-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-016-localize-media-latency", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["get_service", "get_slo_status", "get_ticket", "list_alerts", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_localize_media_latency", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-114 — Localize alarm 9607 (media-service latency)\n\nAlarm 9607 is firing. Work out which service is actually responsible — the service the alarm names is not automatically the culprit — and submit your finding under scope '9607' with the evidence you relied on.", "rubric": {"checks": {"correctness": ["diagnosis_submitted", "service_localized", "fault_confirmed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "efficient_investigation", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-016-localize-media-latency.py"}, "task_id": "dob100-016-localize-media-latency", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "consumer_migrated", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "consumer_before_retire", "staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "api_migration", "code_license": "Apache-2.0", "curation_rationale": ["api_migration", "expert", "34-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-017-auth-v1-to-v2", "llm_judge": false, "origin": "curated", "reference_tool_calls": 34, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_api_endpoints", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "shift_endpoint_traffic", "update_ticket"], "source_split": "train", "source_task_id": "tsk_auth_v1_to_v2", "version": "1.0.0"}, "prompt": "ENG-2402 — Migrate /v1/auth traffic to /v2/auth and retire v1\n\nThe /v1/auth API on api-gateway is superseded by /v2/auth. Retire the legacy path and move its production traffic across, without dropping requests. NovaCart's engineering policies are documented in the knowledge base and are not optional; CI enforces some of them and will tell you what it is unhappy about.\n\nDone when: /v2/auth serves 100% of production traffic, /v1/auth is retired in production, and ENG-2402 is closed.", "rubric": {"checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "consumer_migrated", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "consumer_before_retire", "staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-017-auth-v1-to-v2.py"}, "task_id": "dob100-017-auth-v1-to-v2", "task_name": "ENG-2402 — Migrate /v1/auth traffic to /v2/auth and retire v1", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "consumer_migrated", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "consumer_before_retire", "staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "api_migration", "code_license": "Apache-2.0", "curation_rationale": ["api_migration", "expert", "34-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-018-checkout-v1-to-v2", "llm_judge": false, "origin": "curated", "reference_tool_calls": 34, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_api_endpoints", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "shift_endpoint_traffic", "update_ticket"], "source_split": "train", "source_task_id": "tsk_checkout_v1_to_v2", "version": "1.0.0"}, "prompt": "ENG-2403 — Migrate /v1/checkout traffic to /v2/checkout and retire v1\n\nThe /v1/checkout API on api-gateway is superseded by /v2/checkout. Retire the legacy path and move its production traffic across, without dropping requests. NovaCart's engineering policies are documented in the knowledge base and are not optional; CI enforces some of them and will tell you what it is unhappy about.\n\nDone when: /v2/checkout serves 100% of production traffic, /v1/checkout is retired in production, and ENG-2403 is closed.", "rubric": {"checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "consumer_migrated", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "consumer_before_retire", "staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-018-checkout-v1-to-v2.py"}, "task_id": "dob100-018-checkout-v1-to-v2", "task_name": "ENG-2403 — Migrate /v1/checkout traffic to /v2/checkout and retire v1", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "api_migration", "code_license": "Apache-2.0", "curation_rationale": ["api_migration", "hard", "27-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-019-inventory-v1-to-v2", "llm_judge": false, "origin": "curated", "reference_tool_calls": 27, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_api_endpoints", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "shift_endpoint_traffic", "update_ticket"], "source_split": "train", "source_task_id": "tsk_inventory_v1_to_v2", "version": "1.0.0"}, "prompt": "ENG-2406 — Migrate /v1/inventory traffic to /v2/inventory and retire v1\n\nThe /v1/inventory API on api-gateway is superseded by /v2/inventory. Retire the legacy path and move its production traffic across, without dropping requests. NovaCart's engineering policies are documented in the knowledge base and are not optional; CI enforces some of them and will tell you what it is unhappy about.\n\nDone when: /v2/inventory serves 100% of production traffic, /v1/inventory is retired in production, and ENG-2406 is closed.", "rubric": {"checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-019-inventory-v1-to-v2.py"}, "task_id": "dob100-019-inventory-v1-to-v2", "task_name": "ENG-2406 — Migrate /v1/inventory traffic to /v2/inventory and retire v1", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "api_migration", "code_license": "Apache-2.0", "curation_rationale": ["api_migration", "hard", "27-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-020-media-v1-to-v2", "llm_judge": false, "origin": "curated", "reference_tool_calls": 27, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_api_endpoints", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "shift_endpoint_traffic", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_media_v1_to_v2", "version": "1.0.0"}, "prompt": "ENG-2405 — Migrate /v1/media traffic to /v2/media and retire v1\n\nThe /v1/media API on api-gateway is superseded by /v2/media. Retire the legacy path and move its production traffic across, without dropping requests. NovaCart's engineering policies are documented in the knowledge base and are not optional; CI enforces some of them and will tell you what it is unhappy about.\n\nDone when: /v2/media serves 100% of production traffic, /v1/media is retired in production, and ENG-2405 is closed.", "rubric": {"checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-020-media-v1-to-v2.py"}, "task_id": "dob100-020-media-v1-to-v2", "task_name": "ENG-2405 — Migrate /v1/media traffic to /v2/media and retire v1", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "api_migration", "code_license": "Apache-2.0", "curation_rationale": ["api_migration", "hard", "27-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-021-notify-v1-to-v2", "llm_judge": false, "origin": "curated", "reference_tool_calls": 27, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_api_endpoints", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "shift_endpoint_traffic", "update_ticket"], "source_split": "train", "source_task_id": "tsk_notify_v1_to_v2", "version": "1.0.0"}, "prompt": "ENG-2407 — Migrate /v1/notify traffic to /v2/notify and retire v1\n\nThe /v1/notify API on api-gateway is superseded by /v2/notify. Retire the legacy path and move its production traffic across, without dropping requests. NovaCart's engineering policies are documented in the knowledge base and are not optional; CI enforces some of them and will tell you what it is unhappy about.\n\nDone when: /v2/notify serves 100% of production traffic, /v1/notify is retired in production, and ENG-2407 is closed.", "rubric": {"checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-021-notify-v1-to-v2.py"}, "task_id": "dob100-021-notify-v1-to-v2", "task_name": "ENG-2407 — Migrate /v1/notify traffic to /v2/notify and retire v1", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "consumer_migrated", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "consumer_before_retire", "staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "api_migration", "code_license": "Apache-2.0", "curation_rationale": ["api_migration", "expert", "34-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-022-orders-v1-to-v2", "llm_judge": false, "origin": "curated", "reference_tool_calls": 34, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_api_endpoints", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "shift_endpoint_traffic", "update_ticket"], "source_split": "train", "source_task_id": "tsk_orders_v1_to_v2", "version": "1.0.0"}, "prompt": "ENG-2401 — Migrate /v1/orders traffic to /v2/orders and retire v1\n\nThe /v1/orders API on api-gateway is superseded by /v2/orders. Retire the legacy path and move its production traffic across, without dropping requests. NovaCart's engineering policies are documented in the knowledge base and are not optional; CI enforces some of them and will tell you what it is unhappy about.\n\nDone when: /v2/orders serves 100% of production traffic, /v1/orders is retired in production, and ENG-2401 is closed.", "rubric": {"checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "consumer_migrated", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "consumer_before_retire", "staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-022-orders-v1-to-v2.py"}, "task_id": "dob100-022-orders-v1-to-v2", "task_name": "ENG-2401 — Migrate /v1/orders traffic to /v2/orders and retire v1", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "api_migration", "code_license": "Apache-2.0", "curation_rationale": ["api_migration", "hard", "27-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-023-search-v1-to-v2", "llm_judge": false, "origin": "curated", "reference_tool_calls": 27, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_api_endpoints", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "shift_endpoint_traffic", "update_ticket"], "source_split": "train", "source_task_id": "tsk_search_v1_to_v2", "version": "1.0.0"}, "prompt": "ENG-2404 — Migrate /v1/search traffic to /v2/search and retire v1\n\nThe /v1/search API on api-gateway is superseded by /v2/search. Retire the legacy path and move its production traffic across, without dropping requests. NovaCart's engineering policies are documented in the knowledge base and are not optional; CI enforces some of them and will tell you what it is unhappy about.\n\nDone when: /v2/search serves 100% of production traffic, /v1/search is retired in production, and ENG-2404 is closed.", "rubric": {"checks": {"correctness": ["legacy_retired", "replacement_active", "legacy_traffic_drained", "replacement_serving_all_traffic", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["deprecate_before_shift", "staged_traffic_shifts", "drain_before_retire", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-023-search-v1-to-v2.py"}, "task_id": "dob100-023-search-v1-to-v2", "task_name": "ENG-2404 — Migrate /v1/search traffic to /v2/search and retire v1", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_gateway_latency_surge", "key_for_gateway_latency_surge", "attributed_checkout_error_spike", "key_for_checkout_error_spike", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "both media-service replicas are scheduled on node-b3, which reports DiskPressure at 97% of its ephemeral storage, and one replica has already been Evicted for ephemeral-storage", "fault_detected": true, "fault_type": "node_unhealthy", "offending_key": "node-b3", "scope": "media-upload-stalls", "service": "media-service"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "p99 moved when v5.1.0 was promoted; that release opens an upstream connection per request without releasing it, so the leak is in every version at or above v5.1.0 and only a rollback recovers it", "fault_detected": true, "fault_type": "bad_release", "offending_key": "v5.1.0", "scope": "gateway-latency-surge", "service": "api-gateway"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "checkout errors track the instant_refunds rollout rather than any deploy, and disabling the flag returns the error rate to baseline", "fault_detected": true, "fault_type": "feature_flag_regression", "offending_key": "instant_refunds", "scope": "checkout-error-spike", "service": "checkout"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "attribution", "code_license": "Apache-2.0", "curation_rationale": ["attribution", "expert", "13-call-oracle", "curated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-024-attr-three-at-once", "llm_judge": false, "origin": "curated", "reference_tool_calls": 13, "required_tools": ["get_ticket", "k8s_events_list", "k8s_nodes_list", "k8s_pods_list", "list_alerts", "list_deployments", "list_feature_flags", "query_metrics", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_attr_three_at_once", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nOPS-150 — Three alarms in twenty minutes - is this one incident or three?\n\n3 things are wrong at once this morning:\n\n  - media uploads are stalling for customers\n  - api-gateway p99 is 1030ms against a 250ms SLO\n  - checkout error rate is 5.5% against a 1.0% SLO\n\nOps would like to treat this as a single incident with a single cause, because that is one rollback instead of three investigations. Work out whether that is true. Submit one diagnosis per scope listed above — the scope strings are 'media-upload-stalls', 'gateway-latency-surge', 'checkout-error-spike' — each naming the service, the fault type and the specific setting or version responsible.\n\nThey may share a cause and they may not. Deciding that is the task; do not assume either way, and do not attribute a symptom to a fault you happened to find while looking at something else.", "rubric": {"checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_gateway_latency_surge", "key_for_gateway_latency_surge", "attributed_checkout_error_spike", "key_for_checkout_error_spike", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-024-attr-three-at-once.py"}, "task_id": "dob100-024-attr-three-at-once", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_analytics_connection_refused", "key_for_analytics_connection_refused", "attributed_catalog_latency_sawtooth", "key_for_catalog_latency_sawtooth", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "both replicas sit on node-b3, which reports DiskPressure at 97%", "fault_detected": true, "fault_type": "node_unhealthy", "offending_key": "node-b3", "scope": "media-upload-stalls", "service": "media-service"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "the path is REFUSED at the transport layer, so a network policy rather than capacity", "fault_detected": true, "fault_type": "misconfig", "offending_key": "pg-replica", "scope": "analytics-connection-refused", "service": "analytics-worker"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "94% heap, 780ms p99 GC pause, 41 collections a minute", "fault_detected": true, "fault_type": "resource_exhaustion", "offending_key": "heap_used_pct", "scope": "catalog-latency-sawtooth", "service": "catalog"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "attribution", "code_license": "Apache-2.0", "curation_rationale": ["attribution", "expert", "10-call-oracle", "wave-generated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-025-w5-attr-media-analytics-catalog", "llm_judge": false, "origin": "curated", "reference_tool_calls": 10, "required_tools": ["check_network_path", "get_runtime_stats", "get_ticket", "k8s_nodes_list", "k8s_pods_list", "list_alerts", "submit_diagnosis", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_w5_attr_media_analytics_catalog", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nMULTI-7607 — Three alarms at once - one incident or three?\n\n3 things are wrong at once this morning:\n\n  - media uploads are stalling\n  - analytics cannot reach the replica\n  - catalog latency arrives in waves\n\nOps would like to treat this as a single incident with a single cause, because that is one rollback instead of three investigations. Work out whether that is true. Submit one diagnosis per scope listed above — the scope strings are 'media-upload-stalls', 'analytics-connection-refused', 'catalog-latency-sawtooth' — each naming the service, the fault type and the specific setting or version responsible.\n\nThey may share a cause and they may not. Deciding that is the task; do not assume either way, and do not attribute a symptom to a fault you happened to find while looking at something else.", "rubric": {"checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_analytics_connection_refused", "key_for_analytics_connection_refused", "attributed_catalog_latency_sawtooth", "key_for_catalog_latency_sawtooth", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-025-w5-attr-media-analytics-catalog.py"}, "task_id": "dob100-025-w5-attr-media-analytics-catalog", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_analytics_connection_refused", "key_for_analytics_connection_refused", "attributed_payments_error_rate", "key_for_payments_error_rate", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "both replicas sit on node-b3, which reports DiskPressure at 97%", "fault_detected": true, "fault_type": "node_unhealthy", "offending_key": "node-b3", "scope": "media-upload-stalls", "service": "media-service"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "the path is REFUSED at the transport layer, so a network policy rather than capacity", "fault_detected": true, "fault_type": "misconfig", "offending_key": "pg-replica", "scope": "analytics-connection-refused", "service": "analytics-worker"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "the notifications call times out with no retry configured", "fault_detected": true, "fault_type": "missing_retry", "offending_key": "notifications_retry_max_attempts", "scope": "payments-error-rate", "service": "payments"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "attribution", "code_license": "Apache-2.0", "curation_rationale": ["attribution", "expert", "10-call-oracle", "wave-generated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-026-w5-attr-media-analytics-payments", "llm_judge": false, "origin": "curated", "reference_tool_calls": 10, "required_tools": ["check_network_path", "get_ticket", "k8s_nodes_list", "k8s_pods_list", "list_alerts", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_w5_attr_media_analytics_payments", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nMULTI-7608 — Three alarms at once - one incident or three?\n\n3 things are wrong at once this morning:\n\n  - media uploads are stalling\n  - analytics cannot reach the replica\n  - payments errors are 4.2% against 1.0%\n\nOps would like to treat this as a single incident with a single cause, because that is one rollback instead of three investigations. Work out whether that is true. Submit one diagnosis per scope listed above — the scope strings are 'media-upload-stalls', 'analytics-connection-refused', 'payments-error-rate' — each naming the service, the fault type and the specific setting or version responsible.\n\nThey may share a cause and they may not. Deciding that is the task; do not assume either way, and do not attribute a symptom to a fault you happened to find while looking at something else.", "rubric": {"checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_analytics_connection_refused", "key_for_analytics_connection_refused", "attributed_payments_error_rate", "key_for_payments_error_rate", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-026-w5-attr-media-analytics-payments.py"}, "task_id": "dob100-026-w5-attr-media-analytics-payments", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_gateway_latency_surge", "key_for_gateway_latency_surge", "attributed_analytics_connection_refused", "key_for_analytics_connection_refused", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "both replicas sit on node-b3, which reports DiskPressure at 97%", "fault_detected": true, "fault_type": "node_unhealthy", "offending_key": "node-b3", "scope": "media-upload-stalls", "service": "media-service"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "p99 moved when v5.1.0 was promoted; it leaks an upstream connection per request, so only a rollback recovers it", "fault_detected": true, "fault_type": "bad_release", "offending_key": "v5.1.0", "scope": "gateway-latency-surge", "service": "api-gateway"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "the path is REFUSED at the transport layer, so a network policy rather than capacity", "fault_detected": true, "fault_type": "misconfig", "offending_key": "pg-replica", "scope": "analytics-connection-refused", "service": "analytics-worker"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "attribution", "code_license": "Apache-2.0", "curation_rationale": ["attribution", "expert", "10-call-oracle", "wave-generated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-027-w5-attr-media-api-analytics", "llm_judge": false, "origin": "curated", "reference_tool_calls": 10, "required_tools": ["check_network_path", "get_ticket", "k8s_nodes_list", "k8s_pods_list", "list_alerts", "list_deployments", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_w5_attr_media_api_analytics", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nMULTI-7601 — Three alarms at once - one incident or three?\n\n3 things are wrong at once this morning:\n\n  - media uploads are stalling\n  - api-gateway p99 is 1030ms against 250ms\n  - analytics cannot reach the replica\n\nOps would like to treat this as a single incident with a single cause, because that is one rollback instead of three investigations. Work out whether that is true. Submit one diagnosis per scope listed above — the scope strings are 'media-upload-stalls', 'gateway-latency-surge', 'analytics-connection-refused' — each naming the service, the fault type and the specific setting or version responsible.\n\nThey may share a cause and they may not. Deciding that is the task; do not assume either way, and do not attribute a symptom to a fault you happened to find while looking at something else.", "rubric": {"checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_gateway_latency_surge", "key_for_gateway_latency_surge", "attributed_analytics_connection_refused", "key_for_analytics_connection_refused", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-027-w5-attr-media-api-analytics.py"}, "task_id": "dob100-027-w5-attr-media-api-analytics", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_gateway_latency_surge", "key_for_gateway_latency_surge", "attributed_catalog_latency_sawtooth", "key_for_catalog_latency_sawtooth", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "both replicas sit on node-b3, which reports DiskPressure at 97%", "fault_detected": true, "fault_type": "node_unhealthy", "offending_key": "node-b3", "scope": "media-upload-stalls", "service": "media-service"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "p99 moved when v5.1.0 was promoted; it leaks an upstream connection per request, so only a rollback recovers it", "fault_detected": true, "fault_type": "bad_release", "offending_key": "v5.1.0", "scope": "gateway-latency-surge", "service": "api-gateway"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "94% heap, 780ms p99 GC pause, 41 collections a minute", "fault_detected": true, "fault_type": "resource_exhaustion", "offending_key": "heap_used_pct", "scope": "catalog-latency-sawtooth", "service": "catalog"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "attribution", "code_license": "Apache-2.0", "curation_rationale": ["attribution", "expert", "10-call-oracle", "wave-generated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-028-w5-attr-media-api-catalog", "llm_judge": false, "origin": "curated", "reference_tool_calls": 10, "required_tools": ["get_runtime_stats", "get_ticket", "k8s_nodes_list", "k8s_pods_list", "list_alerts", "list_deployments", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_w5_attr_media_api_catalog", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nMULTI-7602 — Three alarms at once - one incident or three?\n\n3 things are wrong at once this morning:\n\n  - media uploads are stalling\n  - api-gateway p99 is 1030ms against 250ms\n  - catalog latency arrives in waves\n\nOps would like to treat this as a single incident with a single cause, because that is one rollback instead of three investigations. Work out whether that is true. Submit one diagnosis per scope listed above — the scope strings are 'media-upload-stalls', 'gateway-latency-surge', 'catalog-latency-sawtooth' — each naming the service, the fault type and the specific setting or version responsible.\n\nThey may share a cause and they may not. Deciding that is the task; do not assume either way, and do not attribute a symptom to a fault you happened to find while looking at something else.", "rubric": {"checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_gateway_latency_surge", "key_for_gateway_latency_surge", "attributed_catalog_latency_sawtooth", "key_for_catalog_latency_sawtooth", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-028-w5-attr-media-api-catalog.py"}, "task_id": "dob100-028-w5-attr-media-api-catalog", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_gateway_latency_surge", "key_for_gateway_latency_surge", "attributed_checkout_error_spike", "key_for_checkout_error_spike", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "both replicas sit on node-b3, which reports DiskPressure at 97%", "fault_detected": true, "fault_type": "node_unhealthy", "offending_key": "node-b3", "scope": "media-upload-stalls", "service": "media-service"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "p99 moved when v5.1.0 was promoted; it leaks an upstream connection per request, so only a rollback recovers it", "fault_detected": true, "fault_type": "bad_release", "offending_key": "v5.1.0", "scope": "gateway-latency-surge", "service": "api-gateway"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "checkout errors track the instant_refunds rollout rather than any deploy", "fault_detected": true, "fault_type": "feature_flag_regression", "offending_key": "instant_refunds", "scope": "checkout-error-spike", "service": "checkout"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "attribution", "code_license": "Apache-2.0", "curation_rationale": ["attribution", "expert", "10-call-oracle", "wave-generated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-029-w5-attr-media-api-checkout", "llm_judge": false, "origin": "curated", "reference_tool_calls": 10, "required_tools": ["get_ticket", "k8s_nodes_list", "k8s_pods_list", "list_alerts", "list_deployments", "list_feature_flags", "submit_diagnosis", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_w5_attr_media_api_checkout", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nMULTI-7600 — Three alarms at once - one incident or three?\n\n3 things are wrong at once this morning:\n\n  - media uploads are stalling\n  - api-gateway p99 is 1030ms against 250ms\n  - checkout errors are 5.5% against 1.0%\n\nOps would like to treat this as a single incident with a single cause, because that is one rollback instead of three investigations. Work out whether that is true. Submit one diagnosis per scope listed above — the scope strings are 'media-upload-stalls', 'gateway-latency-surge', 'checkout-error-spike' — each naming the service, the fault type and the specific setting or version responsible.\n\nThey may share a cause and they may not. Deciding that is the task; do not assume either way, and do not attribute a symptom to a fault you happened to find while looking at something else.", "rubric": {"checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_gateway_latency_surge", "key_for_gateway_latency_surge", "attributed_checkout_error_spike", "key_for_checkout_error_spike", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-029-w5-attr-media-api-checkout.py"}, "task_id": "dob100-029-w5-attr-media-api-checkout", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_gateway_latency_surge", "key_for_gateway_latency_surge", "attributed_payments_error_rate", "key_for_payments_error_rate", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"evidence": "both replicas sit on node-b3, which reports DiskPressure at 97%", "fault_detected": true, "fault_type": "node_unhealthy", "offending_key": "node-b3", "scope": "media-upload-stalls", "service": "media-service"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "p99 moved when v5.1.0 was promoted; it leaks an upstream connection per request, so only a rollback recovers it", "fault_detected": true, "fault_type": "bad_release", "offending_key": "v5.1.0", "scope": "gateway-latency-surge", "service": "api-gateway"}, "tool": "submit_diagnosis"}, {"arguments": {"evidence": "the notifications call times out with no retry configured", "fault_detected": true, "fault_type": "missing_retry", "offending_key": "notifications_retry_max_attempts", "scope": "payments-error-rate", "service": "payments"}, "tool": "submit_diagnosis"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "attribution", "code_license": "Apache-2.0", "curation_rationale": ["attribution", "expert", "10-call-oracle", "wave-generated", "hardest-of-category-quota", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-030-w5-attr-media-api-payments", "llm_judge": false, "origin": "curated", "reference_tool_calls": 10, "required_tools": ["get_ticket", "k8s_nodes_list", "k8s_pods_list", "list_alerts", "list_deployments", "search_logs", "submit_diagnosis", "update_ticket"], "source_split": "train", "source_task_id": "tsk_w5_attr_media_api_payments", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).\n\nMULTI-7603 — Three alarms at once - one incident or three?\n\n3 things are wrong at once this morning:\n\n  - media uploads are stalling\n  - api-gateway p99 is 1030ms against 250ms\n  - payments errors are 4.2% against 1.0%\n\nOps would like to treat this as a single incident with a single cause, because that is one rollback instead of three investigations. Work out whether that is true. Submit one diagnosis per scope listed above — the scope strings are 'media-upload-stalls', 'gateway-latency-surge', 'payments-error-rate' — each naming the service, the fault type and the specific setting or version responsible.\n\nThey may share a cause and they may not. Deciding that is the task; do not assume either way, and do not attribute a symptom to a fault you happened to find while looking at something else.", "rubric": {"checks": {"correctness": ["attributed_media_upload_stalls", "key_for_media_upload_stalls", "attributed_gateway_latency_surge", "key_for_gateway_latency_surge", "attributed_payments_error_rate", "key_for_payments_error_rate", "causes_are_distinct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["evidence_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-030-w5-attr-media-api-payments.py"}, "task_id": "dob100-030-w5-attr-media-api-payments", "task_name": "This is an investigation, not a change. Do NOT modify production state: no pull requests, deploys, flag toggles, migrations or resolutions. Investigate with the read tools and report what you find with submit_diagnosis(...).", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["implementation_written", "visible_tests_pass", "hidden_tests_pass", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["executed_before_finishing"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "code_implementation", "code_license": "Apache-2.0", "curation_rationale": ["code_implementation", "hard", "5-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-031-impl-backoff", "llm_judge": false, "origin": "curated", "reference_tool_calls": 5, "required_tools": ["get_ticket", "read_exercise", "run_exercise_tests", "update_ticket", "write_implementation"], "source_split": "train", "source_task_id": "tsk_impl_backoff", "version": "1.0.0"}, "prompt": "OPS-140 — Implement the payments retry backoff schedule\n\nsrc/payments/backoff.py is not implemented. Read the specification with read_exercise('src/payments/backoff.py'), write your implementation with write_implementation, and run it with run_exercise_tests until the visible tests pass.\n\nThe visible tests are not the whole specification. Other tests run at grading time and are not shown to you, so satisfying what you can see is necessary and not sufficient — read the specification and implement what it actually says.", "rubric": {"checks": {"correctness": ["implementation_written", "visible_tests_pass", "hidden_tests_pass", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["executed_before_finishing"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-031-impl-backoff.py"}, "task_id": "dob100-031-impl-backoff", "task_name": "OPS-140 — Implement the payments retry backoff schedule", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["implementation_written", "visible_tests_pass", "hidden_tests_pass", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["executed_before_finishing"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "code_implementation", "code_license": "Apache-2.0", "curation_rationale": ["code_implementation", "hard", "5-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-032-impl-cachekey", "llm_judge": false, "origin": "curated", "reference_tool_calls": 5, "required_tools": ["get_ticket", "read_exercise", "run_exercise_tests", "update_ticket", "write_implementation"], "source_split": "train", "source_task_id": "tsk_impl_cachekey", "version": "1.0.0"}, "prompt": "OPS-142 — Implement the search cache key\n\nsrc/search/cache_key.py is not implemented. Read the specification with read_exercise('src/search/cache_key.py'), write your implementation with write_implementation, and run it with run_exercise_tests until the visible tests pass.\n\nThe visible tests are not the whole specification. Other tests run at grading time and are not shown to you, so satisfying what you can see is necessary and not sufficient — read the specification and implement what it actually says.", "rubric": {"checks": {"correctness": ["implementation_written", "visible_tests_pass", "hidden_tests_pass", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["executed_before_finishing"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-032-impl-cachekey.py"}, "task_id": "dob100-032-impl-cachekey", "task_name": "OPS-142 — Implement the search cache key", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["implementation_written", "visible_tests_pass", "hidden_tests_pass", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["executed_before_finishing"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "code_implementation", "code_license": "Apache-2.0", "curation_rationale": ["code_implementation", "medium", "5-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-033-impl-chunk", "llm_judge": false, "origin": "curated", "reference_tool_calls": 5, "required_tools": ["get_ticket", "read_exercise", "run_exercise_tests", "update_ticket", "write_implementation"], "source_split": "train", "source_task_id": "tsk_impl_chunk", "version": "1.0.0"}, "prompt": "OPS-141 — Implement settlement batch chunking\n\nsrc/payments/chunking.py is not implemented. Read the specification with read_exercise('src/payments/chunking.py'), write your implementation with write_implementation, and run it with run_exercise_tests until the visible tests pass.\n\nThe visible tests are not the whole specification. Other tests run at grading time and are not shown to you, so satisfying what you can see is necessary and not sufficient — read the specification and implement what it actually says.", "rubric": {"checks": {"correctness": ["implementation_written", "visible_tests_pass", "hidden_tests_pass", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["executed_before_finishing"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-033-impl-chunk.py"}, "task_id": "dob100-033-impl-chunk", "task_name": "OPS-141 — Implement settlement batch chunking", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["implementation_written", "visible_tests_pass", "hidden_tests_pass", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["executed_before_finishing"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "code_implementation", "code_license": "Apache-2.0", "curation_rationale": ["code_implementation", "expert", "5-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-034-impl-ratelimit", "llm_judge": false, "origin": "curated", "reference_tool_calls": 5, "required_tools": ["get_ticket", "read_exercise", "run_exercise_tests", "update_ticket", "write_implementation"], "source_split": "heldout", "source_task_id": "tsk_impl_ratelimit", "version": "1.0.0"}, "prompt": "OPS-143 — Implement per-client rate limiting at the edge\n\nsrc/gateway/token_bucket.py is not implemented. Read the specification with read_exercise('src/gateway/token_bucket.py'), write your implementation with write_implementation, and run it with run_exercise_tests until the visible tests pass.\n\nThe visible tests are not the whole specification. Other tests run at grading time and are not shown to you, so satisfying what you can see is necessary and not sufficient — read the specification and implement what it actually says.", "rubric": {"checks": {"correctness": ["implementation_written", "visible_tests_pass", "hidden_tests_pass", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["executed_before_finishing"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-034-impl-ratelimit.py"}, "task_id": "dob100-034-impl-ratelimit", "task_name": "OPS-143 — Implement per-client rate limiting at the edge", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["every_match_transitioned", "nothing_extra_transitioned", "outcome_recorded", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "cross_system", "code_license": "Apache-2.0", "curation_rationale": ["cross_system", "hard", "5-call-oracle", "ported", "hardest-of-category-quota", "controls-rejected:wrong_source,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-035-port-close-backlog-issues", "llm_judge": false, "origin": "ported", "reference_tool_calls": 5, "required_tools": ["get_ticket", "jira_search", "jira_transition_issue", "update_ticket"], "source_split": "train", "source_task_id": "tsk_port_close_backlog_issues", "version": "1.0.0"}, "prompt": "ENG-2805 — Empty the Backlog before sprint planning\n\nSprint planning wants the Backlog emptied: move every issue currently in Backlog to Done with a resolution recording why. Issues in any other state are somebody's work in flight and must not be touched.", "rubric": {"checks": {"correctness": ["every_match_transitioned", "nothing_extra_transitioned", "outcome_recorded", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-035-port-close-backlog-issues.py"}, "task_id": "dob100-035-port-close-backlog-issues", "task_name": "ENG-2805 — Empty the Backlog before sprint planning", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["every_match_transitioned", "nothing_extra_transitioned", "outcome_recorded", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "cross_system", "code_license": "Apache-2.0", "curation_rationale": ["cross_system", "expert", "4-call-oracle", "ported", "hardest-of-category-quota", "controls-rejected:wrong_source,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-036-port-close-blocked-issues", "llm_judge": false, "origin": "ported", "reference_tool_calls": 4, "required_tools": ["get_ticket", "jira_search", "jira_transition_issue", "update_ticket"], "source_split": "train", "source_task_id": "tsk_port_close_blocked_issues", "version": "1.0.0"}, "prompt": "ENG-2806 — Clear the Blocked column\n\nAnything sitting in Blocked has been superseded by work already in flight. Move every Blocked issue to Done, recording a resolution that says so. Nothing else moves - In Progress issues are live and Backlog is next sprint's problem.", "rubric": {"checks": {"correctness": ["every_match_transitioned", "nothing_extra_transitioned", "outcome_recorded", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-036-port-close-blocked-issues.py"}, "task_id": "dob100-036-port-close-blocked-issues", "task_name": "ENG-2806 — Clear the Blocked column", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["every_match_copied", "nothing_extra_copied", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth", "announced_the_mirror"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "cross_system", "code_license": "Apache-2.0", "curation_rationale": ["cross_system", "hard", "10-call-oracle", "ported", "hardest-of-category-quota", "controls-rejected:naive,wrong_source,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-037-port-copy-priority-issues", "llm_judge": false, "origin": "ported", "reference_tool_calls": 10, "required_tools": ["create_ticket", "get_ticket", "github_list_issues", "post_message", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_port_copy_priority_issues", "version": "1.0.0"}, "prompt": "ENG-2803 — Mirror priority GitHub issues onto the board\n\nEvery OPEN issue carrying the 'priority' label needs a matching ticket in our own tracker so it appears on the board. Create one ticket per such issue, citing the issue number it came from in the description. Issues that are closed, or that lack the label, must not be copied. Tell #eng once they are mirrored, so nobody picks one up twice.", "rubric": {"checks": {"correctness": ["every_match_copied", "nothing_extra_copied", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth", "announced_the_mirror"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-037-port-copy-priority-issues.py"}, "task_id": "dob100-037-port-copy-priority-issues", "task_name": "ENG-2803 — Mirror priority GitHub issues onto the board", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["answer_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"answer": "6", "assumptions": "Treated Highest as above High, so both count; Medium and Low do not.", "question_id": "Q-PORT-INFLIGHT", "sources": ["jira_issues"]}, "tool": "submit_answer"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "cross_system", "code_license": "Apache-2.0", "curation_rationale": ["cross_system", "hard", "4-call-oracle", "ported", "hardest-of-category-quota", "controls-rejected:wrong_source,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-038-port-count-inflight-work", "llm_judge": false, "origin": "ported", "reference_tool_calls": 4, "required_tools": ["get_ticket", "jira_search", "submit_answer", "update_ticket"], "source_split": "train", "source_task_id": "tsk_port_count_inflight_work", "version": "1.0.0"}, "prompt": "ENG-2808 — How much high-priority work is in flight?\n\nHow many issues are In Progress at High priority or above? Highest counts as above High. Submit the number with submit_answer(question_id='Q-PORT-INFLIGHT', answer=..., sources=[...]).", "rubric": {"checks": {"correctness": ["answer_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-038-port-count-inflight-work.py"}, "task_id": "dob100-038-port-count-inflight-work", "task_name": "ENG-2808 — How much high-priority work is in flight?", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["answer_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"answer": "2", "assumptions": "Counted tickets with no assignee that are not in Done; an unassigned ticket that is already finished does not need an owner.", "question_id": "Q-PORT-UNOWNED", "sources": ["jira_issues"]}, "tool": "submit_answer"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "cross_system", "code_license": "Apache-2.0", "curation_rationale": ["cross_system", "hard", "4-call-oracle", "ported", "hardest-of-category-quota", "controls-rejected:wrong_source,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-039-port-count-unowned-work", "llm_judge": false, "origin": "ported", "reference_tool_calls": 4, "required_tools": ["get_ticket", "jira_search", "submit_answer", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_port_count_unowned_work", "version": "1.0.0"}, "prompt": "ENG-2810 — How much unassigned work is still live?\n\nPlanning wants to know how much live work has nobody on it: how many tickets are unassigned and not already finished? Submit the number with submit_answer(question_id='Q-PORT-UNOWNED', answer=..., sources=[...]), and state in your assumptions how you treated finished tickets - an unassigned ticket that is already done needs no owner, and counting it overstates the problem.", "rubric": {"checks": {"correctness": ["answer_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-039-port-count-unowned-work.py"}, "task_id": "dob100-039-port-count-unowned-work", "task_name": "ENG-2810 — How much unassigned work is still live?", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["every_match_transitioned", "nothing_extra_transitioned", "outcome_recorded", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "cross_system", "code_license": "Apache-2.0", "curation_rationale": ["cross_system", "expert", "5-call-oracle", "ported", "hardest-of-category-quota", "controls-rejected:wrong_source,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-040-port-dedupe-linked-issues", "llm_judge": false, "origin": "ported", "reference_tool_calls": 5, "required_tools": ["get_ticket", "jira_transition_issue", "list_issue_links", "update_ticket"], "source_split": "train", "source_task_id": "tsk_port_dedupe_linked_issues", "version": "1.0.0"}, "prompt": "ENG-2809 — Close the tickets that duplicate other trackers\n\nSome tickets on our board are duplicates of work already tracked in another system, and we keep re-triaging them. Close every ticket that is recorded as duplicating an issue in another tracker, with a resolution that says so. A ticket that merely looks similar is not a duplicate - only the recorded links count, and anything without one stays exactly as it is.", "rubric": {"checks": {"correctness": ["every_match_transitioned", "nothing_extra_transitioned", "outcome_recorded", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-040-port-dedupe-linked-issues.py"}, "task_id": "dob100-040-port-dedupe-linked-issues", "task_name": "ENG-2809 — Close the tickets that duplicate other trackers", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["answer_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"answer": "Priya Nair", "assumptions": "Resolved the gateway to the name the paging system uses, took the escalation policy attached to that service, and read the person holding that policy's schedule on the most recent day.", "question_id": "Q-PORT-ONCALL", "sources": ["pd_oncall"]}, "tool": "submit_answer"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "cross_system", "code_license": "Apache-2.0", "curation_rationale": ["cross_system", "expert", "4-call-oracle", "ported", "hardest-of-category-quota", "controls-rejected:wrong_source,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-041-port-escalate-to-gateway-oncall", "llm_judge": false, "origin": "ported", "reference_tool_calls": 4, "required_tools": ["get_ticket", "pd_list_oncalls", "submit_answer", "update_ticket"], "source_split": "train", "source_task_id": "tsk_port_escalate_to_gateway_oncall", "version": "1.0.0"}, "prompt": "ENG-2812 — Who do we page for the gateway right now?\n\nThe gateway is in trouble and we are about to page. Name the person who is actually on call for it right now, and submit it with submit_answer(question_id='Q-PORT-ONCALL', answer=..., sources=[...]). Paging the wrong person costs us the incident, so go through the escalation policy that covers the gateway rather than whoever is most visible on the schedule - and note that the paging system does not call the gateway what we call it.", "rubric": {"checks": {"correctness": ["answer_correct", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-041-port-escalate-to-gateway-oncall.py"}, "task_id": "dob100-041-port-escalate-to-gateway-oncall", "task_name": "ENG-2812 — Who do we page for the gateway right now?", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["every_match_reported", "nothing_extra_reported", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "cross_system", "code_license": "Apache-2.0", "curation_rationale": ["cross_system", "hard", "4-call-oracle", "ported", "hardest-of-category-quota", "controls-rejected:naive,wrong_source,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-042-port-high-priority-since", "llm_judge": false, "origin": "ported", "reference_tool_calls": 4, "required_tools": ["get_ticket", "github_list_issues", "post_message", "update_ticket"], "source_split": "train", "source_task_id": "tsk_port_high_priority_since", "version": "1.0.0"}, "prompt": "ENG-2802 — Chase the recent high-priority issues in #eng\n\nPost one message to #eng naming every issue that is OPEN, carries the 'priority' label, AND was created on day 410 or later. Reference each by number and ask whether anyone is still on it. Issues that miss any one of those three conditions must not appear - the point of the message is that everything in it needs an owner today.", "rubric": {"checks": {"correctness": ["every_match_reported", "nothing_extra_reported", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["read_the_source_of_truth"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-042-port-high-priority-since.py"}, "task_id": "dob100-042-port-high-priority-since", "task_name": "ENG-2802 — Chase the recent high-priority issues in #eng", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "error_rate_reduction", "code_license": "Apache-2.0", "curation_rationale": ["error_rate_reduction", "medium", "14-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-043-analytics-batch-size", "llm_judge": false, "origin": "curated", "reference_tool_calls": 14, "required_tools": ["deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "query_metrics", "run_ci", "search_docs", "search_logs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_analytics_batch_size", "version": "1.0.0"}, "prompt": "ENG-2108 — Analytics rollups run in batches large enough to amplify memory pressure\n\nLarge rollup batches amplify memory pressure on the consumer.\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, ENG-2108 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-043-analytics-batch-size.py"}, "task_id": "dob100-043-analytics-batch-size", "task_name": "ENG-2108 — Analytics rollups run in batches large enough to amplify memory pressure", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "error_rate_reduction", "code_license": "Apache-2.0", "curation_rationale": ["error_rate_reduction", "medium", "18-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-044-analytics-prefetch", "llm_judge": false, "origin": "curated", "reference_tool_calls": 18, "required_tools": ["acknowledge_alert", "deploy_service", "get_document", "get_ticket", "jira_search", "jira_transition_issue", "merge_pull_request", "open_pull_request", "query_metrics", "resolve_alert", "run_ci", "search_docs", "search_logs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_analytics_prefetch", "version": "1.0.0"}, "prompt": "ENG-2103 — Analytics worker restarting under queue load\n\nAnalytics-worker error_rate_pct is 6.0% against a 2.0% SLO (alarm 9609).\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, error_rate_pct is back inside its SLO, alarm 9609 is resolved, ENG-2103 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-044-analytics-prefetch.py"}, "task_id": "dob100-044-analytics-prefetch", "task_name": "ENG-2103 — Analytics worker restarting under queue load", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "error_rate_reduction", "code_license": "Apache-2.0", "curation_rationale": ["error_rate_reduction", "medium", "16-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-045-checkout-payments-timeout", "llm_judge": false, "origin": "curated", "reference_tool_calls": 16, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "promote_canary", "query_metrics", "run_ci", "search_docs", "search_logs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_checkout_payments_timeout", "version": "1.0.0"}, "prompt": "ENG-2107 — Checkout waits 8s on the payments call\n\nCheckout waits 8s on the payments call, beyond the standard.\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, ENG-2107 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-045-checkout-payments-timeout.py"}, "task_id": "dob100-045-checkout-payments-timeout", "task_name": "ENG-2107 — Checkout waits 8s on the payments call", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "incident_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "error_rate_reduction", "code_license": "Apache-2.0", "curation_rationale": ["error_rate_reduction", "hard", "19-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-046-inventory-pool", "llm_judge": false, "origin": "curated", "reference_tool_calls": 19, "required_tools": ["acknowledge_alert", "deploy_service", "get_document", "get_ticket", "jira_search", "jira_transition_issue", "merge_pull_request", "open_pull_request", "query_metrics", "resolve_alert", "run_ci", "search_docs", "search_logs", "update_incident", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_inventory_pool", "version": "1.0.0"}, "prompt": "ENG-2102 — Inventory reservations failing under peak traffic\n\nInventory error_rate_pct is 4.7% against a 1.0% SLO (alarm 9606) and incident 9703 is open.\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, error_rate_pct is back inside its SLO, alarm 9606 is resolved, incident 9703 is resolved, ENG-2102 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "incident_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-046-inventory-pool.py"}, "task_id": "dob100-046-inventory-pool", "task_name": "ENG-2102 — Inventory reservations failing under peak traffic", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "error_rate_reduction", "code_license": "Apache-2.0", "curation_rationale": ["error_rate_reduction", "medium", "17-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-047-notifications-timeout", "llm_judge": false, "origin": "curated", "reference_tool_calls": 17, "required_tools": ["acknowledge_alert", "deploy_service", "get_document", "get_ticket", "jira_search", "jira_transition_issue", "merge_pull_request", "open_pull_request", "query_metrics", "resolve_alert", "run_ci", "search_docs", "search_logs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_notifications_timeout", "version": "1.0.0"}, "prompt": "ENG-2104 — Notification delivery failures from hung SMTP calls\n\nNotifications error_rate_pct is 3.6% against a 1.5% SLO (alarm 9608).\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, error_rate_pct is back inside its SLO, alarm 9608 is resolved, ENG-2104 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-047-notifications-timeout.py"}, "task_id": "dob100-047-notifications-timeout", "task_name": "ENG-2104 — Notification delivery failures from hung SMTP calls", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "error_rate_reduction", "code_license": "Apache-2.0", "curation_rationale": ["error_rate_reduction", "medium", "15-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-048-payments-notify-timeout", "llm_judge": false, "origin": "curated", "reference_tool_calls": 15, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "promote_canary", "query_metrics", "run_ci", "search_docs", "search_logs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_payments_notify_timeout", "version": "1.0.0"}, "prompt": "ENG-2106 — Payments waits 30s on the notifications call\n\nPayments waits 30s on the notifications call, far beyond the standard.\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, ENG-2106 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-048-payments-notify-timeout.py"}, "task_id": "dob100-048-payments-notify-timeout", "task_name": "ENG-2106 — Payments waits 30s on the notifications call", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "error_rate_reduction", "code_license": "Apache-2.0", "curation_rationale": ["error_rate_reduction", "hard", "19-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-049-payments-retry", "llm_judge": false, "origin": "curated", "reference_tool_calls": 19, "required_tools": ["acknowledge_alert", "assess_canary", "deploy_service", "get_document", "get_ticket", "jira_search", "jira_transition_issue", "merge_pull_request", "open_pull_request", "promote_canary", "query_metrics", "resolve_alert", "run_ci", "search_docs", "search_logs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_payments_retry", "version": "1.0.0"}, "prompt": "ENG-2101 — Payments error rate breaching the 1% SLO\n\nPayments error_rate_pct is 4.2% against a 1.0% SLO (alarm 9601).\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, error_rate_pct is back inside its SLO, alarm 9601 is resolved, ENG-2101 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-049-payments-retry.py"}, "task_id": "dob100-049-payments-retry", "task_name": "ENG-2101 — Payments error rate breaching the 1% SLO", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "error_issue_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "error_rate_reduction", "code_license": "Apache-2.0", "curation_rationale": ["error_rate_reduction", "expert", "19-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-050-payments-retry-from-code", "llm_judge": false, "origin": "curated", "reference_tool_calls": 19, "required_tools": ["acknowledge_alert", "assess_canary", "deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "promote_canary", "query_metrics", "resolve_alert", "resolve_error_event", "run_ci", "search_code", "search_docs", "search_logs", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_payments_retry_from_code", "version": "1.0.0"}, "prompt": "ENG-2105 — Payments: eliminate the permanent-failure path on notification timeouts\n\nPayments error_rate_pct is 4.2% (SLO 1.0%) and the error tracker shows 18k events on 'pay-timeout-01'.\n\nNobody has found the cause yet. The change that introduced it is somewhere in the monorepo history.\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, error_rate_pct is back inside its SLO, alarm 9601 is resolved, the error-tracking issue 'pay-timeout-01' is resolved, ENG-2105 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "error_issue_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-050-payments-retry-from-code.py"}, "task_id": "dob100-050-payments-retry-from-code", "task_name": "ENG-2105 — Payments: eliminate the permanent-failure path on notification timeouts", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["flag_removed", "cleanup_shipped_to_production", "ci_all_stages_green", "no_unrelated_flag_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "feature_flag", "code_license": "Apache-2.0", "curation_rationale": ["feature_flag", "medium", "15-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-051-checkout-v2-layout-cleanup", "llm_judge": false, "origin": "curated", "reference_tool_calls": 15, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_feature_flags", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_checkout_v2_layout_cleanup", "version": "1.0.0"}, "prompt": "ENG-2322 — checkout_v2_layout has been fully rolled out for months\n\nThe checkout_v2_layout flag on checkout has been at 100% in production for months and is now just dead weight in the code. Retire it and make the behaviour unconditional. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the flag no longer exists in any environment, the change is live in production, and ENG-2322 is closed.", "rubric": {"checks": {"correctness": ["flag_removed", "cleanup_shipped_to_production", "ci_all_stages_green", "no_unrelated_flag_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-051-checkout-v2-layout-cleanup.py"}, "task_id": "dob100-051-checkout-v2-layout-cleanup", "task_name": "ENG-2322 — checkout_v2_layout has been fully rolled out for months", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["module_live_in_production", "flag_enabled_at_target_rollout", "flag_defined_in_staging", "ci_all_stages_green", "no_unrelated_flag_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["code_before_flag_enable", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "feature_flag", "code_license": "Apache-2.0", "curation_rationale": ["feature_flag", "hard", "15-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-052-express-checkout", "llm_judge": false, "origin": "curated", "reference_tool_calls": 15, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "set_feature_flag", "update_ticket"], "source_split": "train", "source_task_id": "tsk_express_checkout", "version": "1.0.0"}, "prompt": "ENG-2301 — Ship express checkout behind a feature flag at 10%\n\nImplement the express_checkout module in the checkout service, gated behind a NEW feature flag named 'express_checkout', and roll it out to 10% of production traffic.\n\nThis is a user-facing change, so it ships behind a flag named 'express_checkout'. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the express_checkout module is live in production, 'express_checkout' is serving 10% of production traffic, and ENG-2301 is closed.", "rubric": {"checks": {"correctness": ["module_live_in_production", "flag_enabled_at_target_rollout", "flag_defined_in_staging", "ci_all_stages_green", "no_unrelated_flag_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["code_before_flag_enable", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-052-express-checkout.py"}, "task_id": "dob100-052-express-checkout", "task_name": "ENG-2301 — Ship express checkout behind a feature flag at 10%", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["flag_disabled_in_production", "staging_flag_untouched", "metric_within_slo", "alarm_resolved", "incident_resolved", "no_unrelated_incident_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "no_alarming_deploys"], "quality": ["status_update_posted", "public_status_published", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "feature_flag", "code_license": "Apache-2.0", "curation_rationale": ["feature_flag", "medium", "12-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-053-instant-refunds-killswitch", "llm_judge": false, "origin": "curated", "reference_tool_calls": 12, "required_tools": ["acknowledge_alert", "get_document", "get_ticket", "post_message", "publish_status_update", "query_metrics", "resolve_alert", "search_docs", "search_logs", "set_feature_flag", "update_incident", "update_ticket"], "source_split": "train", "source_task_id": "tsk_instant_refunds_killswitch", "version": "1.0.0"}, "prompt": "ENG-2311 — Checkout error spike since the instant_refunds ramp\n\nIncident 9702: the checkout error rate spiked immediately after the instant_refunds feature flag ramped in production. Staging is unaffected and must be left alone.\n\nMitigate it and see the incident through. NovaCart's incident policy is documented in the knowledge base and is not optional.\n\nDone when: checkout is back inside its SLO, alarm 9603 and incident 9702 are resolved, the team and our customers have been told what happened, and ENG-2311 is closed.", "rubric": {"checks": {"correctness": ["flag_disabled_in_production", "staging_flag_untouched", "metric_within_slo", "alarm_resolved", "incident_resolved", "no_unrelated_incident_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "no_alarming_deploys"], "quality": ["status_update_posted", "public_status_published", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-053-instant-refunds-killswitch.py"}, "task_id": "dob100-053-instant-refunds-killswitch", "task_name": "ENG-2311 — Checkout error spike since the instant_refunds ramp", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["flag_removed", "cleanup_shipped_to_production", "ci_all_stages_green", "no_unrelated_flag_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "feature_flag", "code_license": "Apache-2.0", "curation_rationale": ["feature_flag", "medium", "13-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-054-legacy-price-rounding-cleanup", "llm_judge": false, "origin": "curated", "reference_tool_calls": 13, "required_tools": ["deploy_service", "get_document", "get_ticket", "list_feature_flags", "merge_pull_request", "open_pull_request", "run_ci", "search_docs", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_legacy_price_rounding_cleanup", "version": "1.0.0"}, "prompt": "ENG-2321 — legacy_price_rounding has been fully rolled out for months\n\nThe legacy_price_rounding flag on catalog has been at 100% in production for months and is now just dead weight in the code. Retire it and make the behaviour unconditional. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the flag no longer exists in any environment, the change is live in production, and ENG-2321 is closed.", "rubric": {"checks": {"correctness": ["flag_removed", "cleanup_shipped_to_production", "ci_all_stages_green", "no_unrelated_flag_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-054-legacy-price-rounding-cleanup.py"}, "task_id": "dob100-054-legacy-price-rounding-cleanup", "task_name": "ENG-2321 — legacy_price_rounding has been fully rolled out for months", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["module_live_in_production", "flag_enabled_at_target_rollout", "flag_defined_in_staging", "ci_all_stages_green", "no_unrelated_flag_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["code_before_flag_enable", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "feature_flag", "code_license": "Apache-2.0", "curation_rationale": ["feature_flag", "medium", "12-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-055-media-webp", "llm_judge": false, "origin": "curated", "reference_tool_calls": 12, "required_tools": ["deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "run_ci", "search_docs", "set_feature_flag", "update_ticket"], "source_split": "train", "source_task_id": "tsk_media_webp", "version": "1.0.0"}, "prompt": "ENG-2304 — Ship WebP delivery behind a feature flag\n\nImplement the webp_pipeline module in media-service, gated behind a NEW feature flag named 'webp_delivery', and roll it out to 10% of production traffic.\n\nThis is a user-facing change, so it ships behind a flag named 'webp_delivery'. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the webp_pipeline module is live in production, 'webp_delivery' is serving 10% of production traffic, and ENG-2304 is closed.", "rubric": {"checks": {"correctness": ["module_live_in_production", "flag_enabled_at_target_rollout", "flag_defined_in_staging", "ci_all_stages_green", "no_unrelated_flag_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["code_before_flag_enable", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-055-media-webp.py"}, "task_id": "dob100-055-media-webp", "task_name": "ENG-2304 — Ship WebP delivery behind a feature flag", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["module_live_in_production", "flag_enabled_at_target_rollout", "flag_defined_in_staging", "ci_all_stages_green", "migration_applied_in_production", "no_unrelated_flag_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["code_before_flag_enable", "migration_before_deploy", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "feature_flag", "code_license": "Apache-2.0", "curation_rationale": ["feature_flag", "expert", "20-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-056-saved-carts", "llm_judge": false, "origin": "curated", "reference_tool_calls": 20, "required_tools": ["apply_migration", "assess_canary", "deploy_service", "get_document", "get_ticket", "list_migrations", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "set_feature_flag", "update_ticket"], "source_split": "train", "source_task_id": "tsk_saved_carts", "version": "1.0.0"}, "prompt": "ENG-2303 — Ship saved carts (schema change) behind a feature flag\n\nImplement the saved_carts module in the checkout service, gated behind a NEW feature flag named 'saved_carts', and roll it out to 10% of production traffic.\n\nThis is a user-facing change, so it ships behind a flag named 'saved_carts'. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the saved_carts module is live in production, 'saved_carts' is serving 10% of production traffic, and ENG-2303 is closed.", "rubric": {"checks": {"correctness": ["module_live_in_production", "flag_enabled_at_target_rollout", "flag_defined_in_staging", "ci_all_stages_green", "migration_applied_in_production", "no_unrelated_flag_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["code_before_flag_enable", "migration_before_deploy", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-056-saved-carts.py"}, "task_id": "dob100-056-saved-carts", "task_name": "ENG-2303 — Ship saved carts (schema change) behind a feature flag", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["module_live_in_production", "flag_enabled_at_target_rollout", "flag_defined_in_staging", "ci_all_stages_green", "no_unrelated_flag_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["code_before_flag_enable", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "feature_flag", "code_license": "Apache-2.0", "curation_rationale": ["feature_flag", "medium", "13-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-057-search-autocomplete", "llm_judge": false, "origin": "curated", "reference_tool_calls": 13, "required_tools": ["deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "run_ci", "search_docs", "set_feature_flag", "update_ticket"], "source_split": "train", "source_task_id": "tsk_search_autocomplete", "version": "1.0.0"}, "prompt": "ENG-2302 — Ship search autocomplete behind a feature flag\n\nImplement the autocomplete module in the search service, gated behind a NEW feature flag named 'search_autocomplete', and roll it out to 10% of production traffic.\n\nThis is a user-facing change, so it ships behind a flag named 'search_autocomplete'. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the autocomplete module is live in production, 'search_autocomplete' is serving 10% of production traffic, and ENG-2302 is closed.", "rubric": {"checks": {"correctness": ["module_live_in_production", "flag_enabled_at_target_rollout", "flag_defined_in_staging", "ci_all_stages_green", "no_unrelated_flag_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["code_before_flag_enable", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-057-search-autocomplete.py"}, "task_id": "dob100-057-search-autocomplete", "task_name": "ENG-2302 — Ship search autocomplete behind a feature flag", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["test_fixed_not_quarantined", "three_green_main_runs", "ci_all_stages_green", "no_unrelated_test_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "flaky_test", "code_license": "Apache-2.0", "curation_rationale": ["flaky_test", "medium", "14-call-oracle", "curated", "category-shipped-whole", "controls-rejected:shortcut,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-058-flaky-analytics-rollup", "llm_judge": false, "origin": "curated", "reference_tool_calls": 14, "required_tools": ["get_ci_run", "get_document", "get_ticket", "list_ci_runs", "list_tests", "merge_pull_request", "open_pull_request", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_flaky_analytics_rollup", "version": "1.0.0"}, "prompt": "ENG-2506 — Fix flaky test_rollup_window\n\nanalytics-worker CI is unreliable: test_rollup_window fails intermittently, and people have started re-running the pipeline until it goes green. Make it trustworthy again. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: test_rollup_window is reliably green on the main branch and ENG-2506 is closed.", "rubric": {"checks": {"correctness": ["test_fixed_not_quarantined", "three_green_main_runs", "ci_all_stages_green", "no_unrelated_test_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-058-flaky-analytics-rollup.py"}, "task_id": "dob100-058-flaky-analytics-rollup", "task_name": "ENG-2506 — Fix flaky test_rollup_window", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["test_fixed_not_quarantined", "three_green_main_runs", "ci_all_stages_green", "no_unrelated_test_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "flaky_test", "code_license": "Apache-2.0", "curation_rationale": ["flaky_test", "medium", "14-call-oracle", "curated", "category-shipped-whole", "controls-rejected:shortcut,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-059-flaky-catalog-rounding", "llm_judge": false, "origin": "curated", "reference_tool_calls": 14, "required_tools": ["get_ci_run", "get_document", "get_ticket", "list_ci_runs", "list_tests", "merge_pull_request", "open_pull_request", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_flaky_catalog_rounding", "version": "1.0.0"}, "prompt": "ENG-2502 — Fix flaky test_price_rounding\n\ncatalog CI is unreliable: test_price_rounding fails intermittently, and people have started re-running the pipeline until it goes green. Make it trustworthy again. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: test_price_rounding is reliably green on the main branch and ENG-2502 is closed.", "rubric": {"checks": {"correctness": ["test_fixed_not_quarantined", "three_green_main_runs", "ci_all_stages_green", "no_unrelated_test_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-059-flaky-catalog-rounding.py"}, "task_id": "dob100-059-flaky-catalog-rounding", "task_name": "ENG-2502 — Fix flaky test_price_rounding", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["test_fixed_not_quarantined", "three_green_main_runs", "ci_all_stages_green", "no_unrelated_test_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "flaky_test", "code_license": "Apache-2.0", "curation_rationale": ["flaky_test", "hard", "14-call-oracle", "curated", "category-shipped-whole", "controls-rejected:shortcut,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-060-flaky-checkout-idempotency", "llm_judge": false, "origin": "curated", "reference_tool_calls": 14, "required_tools": ["get_ci_run", "get_document", "get_ticket", "list_ci_runs", "list_tests", "merge_pull_request", "open_pull_request", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_flaky_checkout_idempotency", "version": "1.0.0"}, "prompt": "ENG-2501 — Fix flaky test_checkout_idempotency\n\ncheckout CI is unreliable: test_checkout_idempotency fails intermittently, and people have started re-running the pipeline until it goes green. Make it trustworthy again. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: test_checkout_idempotency is reliably green on the main branch and ENG-2501 is closed.", "rubric": {"checks": {"correctness": ["test_fixed_not_quarantined", "three_green_main_runs", "ci_all_stages_green", "no_unrelated_test_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-060-flaky-checkout-idempotency.py"}, "task_id": "dob100-060-flaky-checkout-idempotency", "task_name": "ENG-2501 — Fix flaky test_checkout_idempotency", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["test_fixed_not_quarantined", "three_green_main_runs", "ci_all_stages_green", "no_unrelated_test_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "flaky_test", "code_license": "Apache-2.0", "curation_rationale": ["flaky_test", "medium", "14-call-oracle", "curated", "category-shipped-whole", "controls-rejected:shortcut,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-061-flaky-gateway-timeout", "llm_judge": false, "origin": "curated", "reference_tool_calls": 14, "required_tools": ["get_ci_run", "get_document", "get_ticket", "list_ci_runs", "list_tests", "merge_pull_request", "open_pull_request", "run_ci", "search_docs", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_flaky_gateway_timeout", "version": "1.0.0"}, "prompt": "ENG-2504 — Fix flaky test_upstream_timeout\n\napi-gateway CI is unreliable: test_upstream_timeout fails intermittently, and people have started re-running the pipeline until it goes green. Make it trustworthy again. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: test_upstream_timeout is reliably green on the main branch and ENG-2504 is closed.", "rubric": {"checks": {"correctness": ["test_fixed_not_quarantined", "three_green_main_runs", "ci_all_stages_green", "no_unrelated_test_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-061-flaky-gateway-timeout.py"}, "task_id": "dob100-061-flaky-gateway-timeout", "task_name": "ENG-2504 — Fix flaky test_upstream_timeout", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["test_fixed_not_quarantined", "three_green_main_runs", "ci_all_stages_green", "no_unrelated_test_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "flaky_test", "code_license": "Apache-2.0", "curation_rationale": ["flaky_test", "hard", "14-call-oracle", "curated", "category-shipped-whole", "controls-rejected:shortcut,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-062-flaky-inventory-race", "llm_judge": false, "origin": "curated", "reference_tool_calls": 14, "required_tools": ["get_ci_run", "get_document", "get_ticket", "list_ci_runs", "list_tests", "merge_pull_request", "open_pull_request", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_flaky_inventory_race", "version": "1.0.0"}, "prompt": "ENG-2503 — Fix flaky test_reservation_race\n\ninventory CI is unreliable: test_reservation_race fails intermittently, and people have started re-running the pipeline until it goes green. Make it trustworthy again. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: test_reservation_race is reliably green on the main branch and ENG-2503 is closed.", "rubric": {"checks": {"correctness": ["test_fixed_not_quarantined", "three_green_main_runs", "ci_all_stages_green", "no_unrelated_test_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-062-flaky-inventory-race.py"}, "task_id": "dob100-062-flaky-inventory-race", "task_name": "ENG-2503 — Fix flaky test_reservation_race", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["test_fixed_not_quarantined", "three_green_main_runs", "ci_all_stages_green", "no_unrelated_test_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "flaky_test", "code_license": "Apache-2.0", "curation_rationale": ["flaky_test", "medium", "14-call-oracle", "curated", "category-shipped-whole", "controls-rejected:shortcut,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-063-flaky-search-index", "llm_judge": false, "origin": "curated", "reference_tool_calls": 14, "required_tools": ["get_ci_run", "get_document", "get_ticket", "list_ci_runs", "list_tests", "merge_pull_request", "open_pull_request", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_flaky_search_index", "version": "1.0.0"}, "prompt": "ENG-2505 — Fix flaky test_index_refresh\n\nsearch CI is unreliable: test_index_refresh fails intermittently, and people have started re-running the pipeline until it goes green. Make it trustworthy again. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: test_index_refresh is reliably green on the main branch and ENG-2505 is closed.", "rubric": {"checks": {"correctness": ["test_fixed_not_quarantined", "three_green_main_runs", "ci_all_stages_green", "no_unrelated_test_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-063-flaky-search-index.py"}, "task_id": "dob100-063-flaky-search-index", "task_name": "ENG-2505 — Fix flaky test_index_refresh", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["page_written", "states_the_node", "states_the_condition", "states_the_service", "states_the_second_node", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "handover", "code_license": "Apache-2.0", "curation_rationale": ["handover", "expert", "6-call-oracle", "ported", "category-shipped-whole", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-064-hand-cluster-runbook", "llm_judge": false, "origin": "ported", "reference_tool_calls": 6, "required_tools": ["get_ticket", "k8s_events_list", "k8s_nodes_list", "k8s_pods_list", "update_ticket", "write_runbook"], "source_split": "train", "source_task_id": "tsk_hand_cluster_runbook", "version": "1.0.0"}, "prompt": "This is an investigation and a write-up, not a change. Do NOT modify production state.\n\nDOC-3002 — Write the cluster-fault runbook\n\nThree separate incidents this month turned out to be the cluster rather than the service, and each took hours because nobody looked below the pod. Write the page that would have saved that time: which node is unhealthy and how, and which service is affected by it.\n\nWrite it with write_runbook(title=..., body=...). It has to be usable by someone who was not here, which means the specific numbers and names, not a summary of the shape of the problem.", "rubric": {"checks": {"correctness": ["page_written", "states_the_node", "states_the_condition", "states_the_service", "states_the_second_node", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-064-hand-cluster-runbook.py"}, "task_id": "dob100-064-hand-cluster-runbook", "task_name": "This is an investigation and a write-up, not a change. Do NOT modify production state.", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["page_written", "states_the_release", "states_the_measurement", "states_the_objective", "states_why_rollforward_fails", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "handover", "code_license": "Apache-2.0", "curation_rationale": ["handover", "hard", "7-call-oracle", "ported", "category-shipped-whole", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-065-hand-gateway-runbook", "llm_judge": false, "origin": "ported", "reference_tool_calls": 7, "required_tools": ["get_slo_status", "get_ticket", "list_deployments", "query_metrics", "search_logs", "update_ticket", "write_runbook"], "source_split": "train", "source_task_id": "tsk_hand_gateway_runbook", "version": "1.0.0"}, "prompt": "This is an investigation and a write-up, not a change. Do NOT modify production state.\n\nDOC-3001 — Write the gateway latency runbook\n\nThe gateway incident is the third of its kind and nobody has written it down. Write the runbook page a new on-call would need: which release introduced it, what the p99 reached against what objective, and why rolling FORWARD does not recover it.\n\nWrite it with write_runbook(title=..., body=...). It has to be usable by someone who was not here, which means the specific numbers and names, not a summary of the shape of the problem.", "rubric": {"checks": {"correctness": ["page_written", "states_the_release", "states_the_measurement", "states_the_objective", "states_why_rollforward_fails", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-065-hand-gateway-runbook.py"}, "task_id": "dob100-065-hand-gateway-runbook", "task_name": "This is an investigation and a write-up, not a change. Do NOT modify production state.", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["service_recovered", "root_cause_written_down", "followup_left_open", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["mitigated_with_the_right_lever", "ack_before_mitigating"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "horizon", "code_license": "Apache-2.0", "curation_rationale": ["horizon", "hard", "7-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-066-hz-checkout-flag-kill", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["acknowledge_alert", "create_ticket", "get_ticket", "list_feature_flags", "query_metrics", "set_feature_flag", "update_ticket"], "source_split": "train", "source_task_id": "tsk_hz_checkout_flag_kill", "version": "1.0.0"}, "prompt": "OPS-171 — Checkout errors - stop the bleeding\n\ncheckout error rate is 5.5% against a 1.0% objective. The errors track the instant_refunds rollout rather than any deploy, so the flag is the fastest way to stop them - but the code behind the flag is still wrong.\n\nStop the bleeding first. Then make sure the thing that actually caused it is written down somewhere the next person will find it - the mitigation is not the fix, and the graph looking healthy is not the same as the defect being gone.", "rubric": {"checks": {"correctness": ["service_recovered", "root_cause_written_down", "followup_left_open", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["mitigated_with_the_right_lever", "ack_before_mitigating"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-066-hz-checkout-flag-kill.py"}, "task_id": "dob100-066-hz-checkout-flag-kill", "task_name": "OPS-171 — Checkout errors - stop the bleeding", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["service_recovered", "root_cause_written_down", "followup_left_open", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["mitigated_with_the_right_lever", "ack_before_mitigating"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "horizon", "code_license": "Apache-2.0", "curation_rationale": ["horizon", "expert", "7-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-067-hz-gateway-leak-rollback", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["acknowledge_alert", "create_ticket", "get_ticket", "list_deployments", "rollback_deployment", "search_logs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_hz_gateway_leak_rollback", "version": "1.0.0"}, "prompt": "OPS-170 — Gateway latency - stop the bleeding\n\napi-gateway p99 is 1030ms against a 250ms objective and has been since v5.1.0 was promoted. That release opens an upstream connection per request without releasing it, so the leak is in every version at or above v5.1.0 - rolling forward does not recover it.\n\nStop the bleeding first. Then make sure the thing that actually caused it is written down somewhere the next person will find it - the mitigation is not the fix, and the graph looking healthy is not the same as the defect being gone.", "rubric": {"checks": {"correctness": ["service_recovered", "root_cause_written_down", "followup_left_open", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["mitigated_with_the_right_lever", "ack_before_mitigating"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-067-hz-gateway-leak-rollback.py"}, "task_id": "dob100-067-hz-gateway-leak-rollback", "task_name": "OPS-170 — Gateway latency - stop the bleeding", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["approval_was_requested", "approval_was_granted", "the_work_landed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["approval_preceded_the_action", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["justification_was_specific", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "human_gated", "code_license": "Apache-2.0", "curation_rationale": ["human_gated", "expert", "18-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-068-gated-rotate-partner-credential", "llm_judge": false, "origin": "curated", "reference_tool_calls": 18, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_approval_policy", "merge_pull_request", "open_pull_request", "promote_canary", "request_approval", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_gated_rotate_partner_credential", "version": "1.0.0"}, "prompt": "SEC-908 — Rotate the exposed partner API credential\n\nThe partner API key was committed to the monorepo, so it must be treated as compromised and rotated in production. Rotation invalidates every existing partner session immediately and cannot be undone - a mistake locks partners out of production until a new key is issued.\n\nNovaCart's engineering policies are documented in the knowledge base and are not optional. Some actions are not yours to take unilaterally.\n\nDone when: checkout is running the rotated key version in production with the sign-off it requires and SEC-908 is closed.", "rubric": {"checks": {"correctness": ["approval_was_requested", "approval_was_granted", "the_work_landed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["approval_preceded_the_action", "staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["justification_was_specific", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-068-gated-rotate-partner-credential.py"}, "task_id": "dob100-068-gated-rotate-partner-credential", "task_name": "SEC-908 — Rotate the exposed partner API credential", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["decision_submitted", "chose_the_root_cause_fix", "read_the_proposals", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["decision_was_read_only"], "quality": ["reasoning_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"answer": "202", "assumptions": "202 fixes the cause. prefetch_count=0 means unlimited prefetch, so the consumer pulls the whole backlog into memory and is killed at 511Mi of a 512Mi limit. 201 raises the limit, which moves the threshold and will fail again on a larger backlog. 203 makes the crashloop quieter without stopping it. 204 disables the feature, which is availability loss dressed as a fix.", "question_id": "J-OOM", "sources": ["remediation_proposals"]}, "tool": "submit_answer"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "judgement", "code_license": "Apache-2.0", "curation_rationale": ["judgement", "expert", "8-call-oracle", "curated", "category-shipped-whole", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-069-judge-analytics-oom", "llm_judge": false, "origin": "curated", "reference_tool_calls": 8, "required_tools": ["get_document", "get_ticket", "k8s_events_list", "k8s_pods_list", "list_remediation_proposals", "search_docs", "submit_answer", "update_ticket"], "source_split": "train", "source_task_id": "tsk_judge_analytics_oom", "version": "1.0.0"}, "prompt": "This is a decision, not a change. Do NOT modify production state.\n\nOPS-302 — Decide the remediation for the analytics crashloop\n\nanalytics-worker is being OOMKilled and is in CrashLoopBackOff. Four remediations have been proposed.\n\nFour people have proposed a remediation. Read the proposals and the evidence, decide which one is actually right, and submit its proposal_id with submit_answer(question_id='J-OOM', answer='<id>'). Several of the proposals will reduce the symptom; only one addresses the cause. In `assumptions`, say why the others are wrong.", "rubric": {"checks": {"correctness": ["decision_submitted", "chose_the_root_cause_fix", "read_the_proposals", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["decision_was_read_only"], "quality": ["reasoning_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-069-judge-analytics-oom.py"}, "task_id": "dob100-069-judge-analytics-oom", "task_name": "This is a decision, not a change. Do NOT modify production state.", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["decision_submitted", "chose_the_root_cause_fix", "read_the_proposals", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["decision_was_read_only"], "quality": ["reasoning_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"answer": "401", "assumptions": "401 is correct for mitigation. The errors correlate 1:1 with the flag, and a flag is a runtime toggle that takes effect immediately - the incident runbook puts mitigation before the durable fix. 402 rolls back a deploy that did not introduce the fault, so it would not help. 403 is the right permanent fix but requires CI, merge, staging and a canary while customers are erroring. 404 changes a timeout unrelated to a null dereference.", "question_id": "J-CHK", "sources": ["remediation_proposals"]}, "tool": "submit_answer"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "judgement", "code_license": "Apache-2.0", "curation_rationale": ["judgement", "hard", "8-call-oracle", "curated", "category-shipped-whole", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-070-judge-checkout-flag", "llm_judge": false, "origin": "curated", "reference_tool_calls": 8, "required_tools": ["get_document", "get_ticket", "list_feature_flags", "list_remediation_proposals", "search_docs", "search_logs", "submit_answer", "update_ticket"], "source_split": "train", "source_task_id": "tsk_judge_checkout_flag", "version": "1.0.0"}, "prompt": "This is a decision, not a change. Do NOT modify production state.\n\nOPS-304 — Decide the remediation for the checkout error spike\n\ncheckout is failing 5.5% of requests and the rate tracks the instant_refunds flag ramp exactly. Four remediations have been proposed and customers are affected now.\n\nFour people have proposed a remediation. Read the proposals and the evidence, decide which one is actually right, and submit its proposal_id with submit_answer(question_id='J-CHK', answer='<id>'). Several of the proposals will reduce the symptom; only one addresses the cause. In `assumptions`, say why the others are wrong.", "rubric": {"checks": {"correctness": ["decision_submitted", "chose_the_root_cause_fix", "read_the_proposals", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["decision_was_read_only"], "quality": ["reasoning_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-070-judge-checkout-flag.py"}, "task_id": "dob100-070-judge-checkout-flag", "task_name": "This is a decision, not a change. Do NOT modify production state.", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["decision_submitted", "chose_the_root_cause_fix", "read_the_proposals", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["decision_was_read_only"], "quality": ["reasoning_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"answer": "303", "assumptions": "303 is correct: p99 moved at the exact moment v5.1.0 was promoted, and that release opens an upstream connection per request without releasing it. Because the leak is in every version at or above v5.1.0, rolling forward does not recover it - only a rollback does. 301 and 302 add capacity to a leak, which delays exhaustion rather than preventing it. 304 raises the SLO, which changes the measurement rather than the system.", "question_id": "J-GW", "sources": ["remediation_proposals"]}, "tool": "submit_answer"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "judgement", "code_license": "Apache-2.0", "curation_rationale": ["judgement", "expert", "7-call-oracle", "curated", "category-shipped-whole", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-071-judge-gateway-latency", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["get_ticket", "k8s_pods_list", "list_deployments", "list_remediation_proposals", "search_logs", "submit_answer", "update_ticket"], "source_split": "train", "source_task_id": "tsk_judge_gateway_latency", "version": "1.0.0"}, "prompt": "This is a decision, not a change. Do NOT modify production state.\n\nOPS-303 — Decide the remediation for the gateway latency surge\n\napi-gateway p99 is 1030ms against a 250ms SLO and has been since v5.1.0 was promoted. Four remediations have been proposed.\n\nFour people have proposed a remediation. Read the proposals and the evidence, decide which one is actually right, and submit its proposal_id with submit_answer(question_id='J-GW', answer='<id>'). Several of the proposals will reduce the symptom; only one addresses the cause. In `assumptions`, say why the others are wrong.", "rubric": {"checks": {"correctness": ["decision_submitted", "chose_the_root_cause_fix", "read_the_proposals", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["decision_was_read_only"], "quality": ["reasoning_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-071-judge-gateway-latency.py"}, "task_id": "dob100-071-judge-gateway-latency", "task_name": "This is a decision, not a change. Do NOT modify production state.", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["decision_submitted", "chose_the_root_cause_fix", "read_the_proposals", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["decision_was_read_only"], "quality": ["reasoning_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"answer": "102", "assumptions": "102 is the only proposal that addresses the mechanism: a single notifications timeout permanently fails the payment because no retry is attempted, and the retry standard requires 3. 101 lengthens the timeout, which reduces how often the symptom appears without removing the permanent failure. 103 makes the call fire-and-forget, which silently drops receipts - it changes behaviour rather than restoring it. 104 scales notifications, which is not the bottleneck: the downstream answers, our caller gives up.", "question_id": "J-PAY", "sources": ["remediation_proposals"]}, "tool": "submit_answer"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "judgement", "code_license": "Apache-2.0", "curation_rationale": ["judgement", "hard", "7-call-oracle", "curated", "category-shipped-whole", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-072-judge-payments-retry", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["get_document", "get_ticket", "list_remediation_proposals", "search_docs", "search_logs", "submit_answer", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_judge_payments_retry", "version": "1.0.0"}, "prompt": "This is a decision, not a change. Do NOT modify production state.\n\nOPS-301 — Decide the remediation for the payments error rate\n\npayments is failing 4.2% of requests against a 1.0% SLO. The team has proposed four remediations in the incident channel and wants a decision.\n\nFour people have proposed a remediation. Read the proposals and the evidence, decide which one is actually right, and submit its proposal_id with submit_answer(question_id='J-PAY', answer='<id>'). Several of the proposals will reduce the symptom; only one addresses the cause. In `assumptions`, say why the others are wrong.", "rubric": {"checks": {"correctness": ["decision_submitted", "chose_the_root_cause_fix", "read_the_proposals", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["decision_was_read_only"], "quality": ["reasoning_recorded", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-072-judge-payments-retry.py"}, "task_id": "dob100-072-judge-payments-retry", "task_name": "This is a decision, not a change. Do NOT modify production state.", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "latency_optimization", "code_license": "Apache-2.0", "curation_rationale": ["latency_optimization", "hard", "19-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-073-catalog-batch-pricing", "llm_judge": false, "origin": "curated", "reference_tool_calls": 19, "required_tools": ["acknowledge_alert", "deploy_service", "get_document", "get_ticket", "get_traffic_stats", "jira_search", "jira_transition_issue", "merge_pull_request", "open_pull_request", "query_metrics", "resolve_alert", "run_ci", "search_docs", "search_logs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_catalog_batch_pricing", "version": "1.0.0"}, "prompt": "ENG-2202 — Catalog pricing p99 regression\n\nCatalog latency_p99_ms is 645ms against a 300ms SLO (alarm 9605).\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, latency_p99_ms is back inside its SLO, alarm 9605 is resolved, ENG-2202 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-073-catalog-batch-pricing.py"}, "task_id": "dob100-073-catalog-batch-pricing", "task_name": "ENG-2202 — Catalog pricing p99 regression", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "latency_optimization", "code_license": "Apache-2.0", "curation_rationale": ["latency_optimization", "medium", "14-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-074-catalog-cache-ttl", "llm_judge": false, "origin": "curated", "reference_tool_calls": 14, "required_tools": ["deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "query_metrics", "run_ci", "search_docs", "search_logs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_catalog_cache_ttl", "version": "1.0.0"}, "prompt": "ENG-2206 — Catalog cache entries expire sooner than the standard allows\n\nCatalog caches entries for only 120s, well under the standard.\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, ENG-2206 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-074-catalog-cache-ttl.py"}, "task_id": "dob100-074-catalog-cache-ttl", "task_name": "ENG-2206 — Catalog cache entries expire sooner than the standard allows", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "latency_optimization", "code_license": "Apache-2.0", "curation_rationale": ["latency_optimization", "expert", "18-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-075-catalog-pricing-from-code", "llm_judge": false, "origin": "curated", "reference_tool_calls": 18, "required_tools": ["acknowledge_alert", "deploy_service", "get_document", "get_ticket", "get_traffic_stats", "merge_pull_request", "open_pull_request", "query_metrics", "resolve_alert", "run_ci", "search_code", "search_docs", "search_logs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_catalog_pricing_from_code", "version": "1.0.0"}, "prompt": "ENG-2204 — Catalog: remove the N+1 pricing query from the hot path\n\nCatalog latency_p99_ms is 645ms (SLO 300ms).\n\nNobody has found the cause yet. The change that introduced it is somewhere in the monorepo history.\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, latency_p99_ms is back inside its SLO, alarm 9605 is resolved, ENG-2204 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-075-catalog-pricing-from-code.py"}, "task_id": "dob100-075-catalog-pricing-from-code", "task_name": "ENG-2204 — Catalog: remove the N+1 pricing query from the hot path", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "latency_optimization", "code_license": "Apache-2.0", "curation_rationale": ["latency_optimization", "hard", "16-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-076-gateway-pool-reuse", "llm_judge": false, "origin": "curated", "reference_tool_calls": 16, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "promote_canary", "query_metrics", "run_ci", "search_docs", "search_logs", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_gateway_pool_reuse", "version": "1.0.0"}, "prompt": "ENG-2205 — API gateway holds a new upstream connection per request\n\nThe gateway opens a new upstream connection per request and never releases it.\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, ENG-2205 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-076-gateway-pool-reuse.py"}, "task_id": "dob100-076-gateway-pool-reuse", "task_name": "ENG-2205 — API gateway holds a new upstream connection per request", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["rolled_back_to_good_version", "bad_deploy_marked_rolled_back", "metric_within_slo", "alarm_resolved", "incident_resolved", "staging_untouched", "no_unrelated_incident_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["rollback_tool_used", "ack_before_resolve"], "quality": ["postmortem_filed", "status_update_posted", "public_status_published", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "latency_optimization", "code_license": "Apache-2.0", "curation_rationale": ["latency_optimization", "hard", "15-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-077-gateway-v510-rollback", "llm_judge": false, "origin": "curated", "reference_tool_calls": 15, "required_tools": ["acknowledge_alert", "create_ticket", "get_document", "get_ticket", "get_traffic_stats", "list_deployments", "post_message", "publish_status_update", "query_metrics", "resolve_alert", "rollback_deployment", "search_docs", "search_logs", "update_incident", "update_ticket"], "source_split": "train", "source_task_id": "tsk_gateway_v510_rollback", "version": "1.0.0"}, "prompt": "ENG-2208 — SEV1: api-gateway latency surge since the v5.1.0 rollout\n\nIncident 9701, SEV1: api-gateway p99 latency jumped immediately after v5.1.0 was promoted in production and has stayed there. Customers are affected right now — stop the bleeding first, then see the incident through. Staging must be left as it is.\n\nNovaCart's incident policy is documented in the knowledge base and is not optional.\n\nDone when: api-gateway latency is back inside its SLO, alarm 9604 and incident 9701 are resolved, the team and our customers have been told what happened, the follow-up work is recorded, and ENG-2208 is closed.", "rubric": {"checks": {"correctness": ["rolled_back_to_good_version", "bad_deploy_marked_rolled_back", "metric_within_slo", "alarm_resolved", "incident_resolved", "staging_untouched", "no_unrelated_incident_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["rollback_tool_used", "ack_before_resolve"], "quality": ["postmortem_filed", "status_update_posted", "public_status_published", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-077-gateway-v510-rollback.py"}, "task_id": "dob100-077-gateway-v510-rollback", "task_name": "ENG-2208 — SEV1: api-gateway latency surge since the v5.1.0 rollout", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "latency_optimization", "code_license": "Apache-2.0", "curation_rationale": ["latency_optimization", "medium", "18-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-078-media-cdn", "llm_judge": false, "origin": "curated", "reference_tool_calls": 18, "required_tools": ["acknowledge_alert", "deploy_service", "get_document", "get_ticket", "get_traffic_stats", "jira_search", "jira_transition_issue", "merge_pull_request", "open_pull_request", "query_metrics", "resolve_alert", "run_ci", "search_docs", "search_logs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_media_cdn", "version": "1.0.0"}, "prompt": "ENG-2203 — Media assets served from origin instead of the CDN\n\nMedia-service latency_p99_ms is 800ms against a 400ms SLO (alarm 9607).\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, latency_p99_ms is back inside its SLO, alarm 9607 is resolved, ENG-2203 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-078-media-cdn.py"}, "task_id": "dob100-078-media-cdn", "task_name": "ENG-2203 — Media assets served from origin instead of the CDN", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "latency_optimization", "code_license": "Apache-2.0", "curation_rationale": ["latency_optimization", "medium", "19-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-079-search-cache", "llm_judge": false, "origin": "curated", "reference_tool_calls": 19, "required_tools": ["acknowledge_alert", "deploy_service", "get_document", "get_ticket", "get_traffic_stats", "jira_search", "jira_transition_issue", "merge_pull_request", "open_pull_request", "query_metrics", "resolve_alert", "run_ci", "search_docs", "search_logs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_search_cache", "version": "1.0.0"}, "prompt": "ENG-2201 — Search p99 latency exceeds the 300ms SLO\n\nSearch latency_p99_ms is 850ms against a 300ms SLO (alarm 9602).\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, latency_p99_ms is back inside its SLO, alarm 9602 is resolved, ENG-2201 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "metric_within_slo", "alarm_resolved", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ack_before_resolve", "staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "jira_twin_resolved", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-079-search-cache.py"}, "task_id": "dob100-079-search-cache", "task_name": "ENG-2201 — Search p99 latency exceeds the 300ms SLO", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "latency_optimization", "code_license": "Apache-2.0", "curation_rationale": ["latency_optimization", "medium", "14-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-080-search-shards", "llm_judge": false, "origin": "curated", "reference_tool_calls": 14, "required_tools": ["deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "query_metrics", "run_ci", "search_docs", "search_logs", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_search_shards", "version": "1.0.0"}, "prompt": "ENG-2207 — Search query fan-out is narrower than the index can support\n\nSearch queries fan out over only 4 shards at 180 rps.\n\nFind the root cause and ship a fix all the way to production. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: the fix is live in production, ENG-2207 is closed.", "rubric": {"checks": {"correctness": ["config_deployed_to_production", "ci_all_stages_green", "no_unrelated_config_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-080-search-shards.py"}, "task_id": "dob100-080-search-shards", "task_name": "ENG-2207 — Search query fan-out is narrower than the index can support", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["module_live_inventory", "ci_stages_inventory", "migration_applied_inventory", "module_live_checkout", "ci_stages_checkout", "module_live_storefront-web", "ci_stages_storefront-web", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_inventory", "pr_linked_checkout", "pr_linked_storefront-web", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "multi_service_rollout", "code_license": "Apache-2.0", "curation_rationale": ["multi_service_rollout", "expert", "31-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-081-backorders", "llm_judge": false, "origin": "curated", "reference_tool_calls": 31, "required_tools": ["apply_migration", "assess_canary", "deploy_service", "get_document", "get_ticket", "list_migrations", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_backorders", "version": "1.0.0"}, "prompt": "ENG-2603 — Roll out backorders across inventory, checkout and storefront-web\n\nShip: module backorder_queue in inventory (needs migration 0034_backorders), module backorder_flow in checkout, module backorder_banner in storefront-web. Each service consumes the one before it, so production must come up in this order: inventory then checkout then storefront-web. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: all of it is live in production in that order and ENG-2603 is closed.", "rubric": {"checks": {"correctness": ["module_live_inventory", "ci_stages_inventory", "migration_applied_inventory", "module_live_checkout", "ci_stages_checkout", "module_live_storefront-web", "ci_stages_storefront-web", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_inventory", "pr_linked_checkout", "pr_linked_storefront-web", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-081-backorders.py"}, "task_id": "dob100-081-backorders", "task_name": "ENG-2603 — Roll out backorders across inventory, checkout and storefront-web", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["module_live_catalog", "ci_stages_catalog", "module_live_checkout", "ci_stages_checkout", "module_live_storefront-web", "ci_stages_storefront-web", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_catalog", "pr_linked_checkout", "pr_linked_storefront-web", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "multi_service_rollout", "code_license": "Apache-2.0", "curation_rationale": ["multi_service_rollout", "expert", "26-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-082-checkout-upsell", "llm_judge": false, "origin": "curated", "reference_tool_calls": 26, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_checkout_upsell", "version": "1.0.0"}, "prompt": "ENG-2607 — Roll out checkout upsell across catalog, checkout and storefront-web\n\nShip: module upsell_slots in catalog, module upsell_flow in checkout, module upsell_widget in storefront-web. Each service consumes the one before it, so production must come up in this order: catalog then checkout then storefront-web. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: all of it is live in production in that order and ENG-2607 is closed.", "rubric": {"checks": {"correctness": ["module_live_catalog", "ci_stages_catalog", "module_live_checkout", "ci_stages_checkout", "module_live_storefront-web", "ci_stages_storefront-web", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_catalog", "pr_linked_checkout", "pr_linked_storefront-web", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-082-checkout-upsell.py"}, "task_id": "dob100-082-checkout-upsell", "task_name": "ENG-2607 — Roll out checkout upsell across catalog, checkout and storefront-web", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["module_live_catalog", "ci_stages_catalog", "migration_applied_catalog", "module_live_checkout", "ci_stages_checkout", "migration_applied_checkout", "module_live_storefront-web", "ci_stages_storefront-web", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_catalog", "pr_linked_checkout", "pr_linked_storefront-web", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "multi_service_rollout", "code_license": "Apache-2.0", "curation_rationale": ["multi_service_rollout", "expert", "33-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-083-loyalty-points", "llm_judge": false, "origin": "curated", "reference_tool_calls": 33, "required_tools": ["apply_migration", "assess_canary", "deploy_service", "get_document", "get_ticket", "list_migrations", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_loyalty_points", "version": "1.0.0"}, "prompt": "ENG-2601 — Roll out loyalty points across catalog, checkout and storefront-web\n\nShip: module loyalty_accrual in catalog (needs migration 0123_loyalty_points), module loyalty_redeem in checkout (needs migration 0088_loyalty_ledger), module loyalty_widget in storefront-web. Each service consumes the one before it, so production must come up in this order: catalog then checkout then storefront-web. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: all of it is live in production in that order and ENG-2601 is closed.", "rubric": {"checks": {"correctness": ["module_live_catalog", "ci_stages_catalog", "migration_applied_catalog", "module_live_checkout", "ci_stages_checkout", "migration_applied_checkout", "module_live_storefront-web", "ci_stages_storefront-web", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_catalog", "pr_linked_checkout", "pr_linked_storefront-web", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-083-loyalty-points.py"}, "task_id": "dob100-083-loyalty-points", "task_name": "ENG-2601 — Roll out loyalty points across catalog, checkout and storefront-web", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["module_live_media-service", "ci_stages_media-service", "module_live_catalog", "ci_stages_catalog", "module_live_storefront-web", "ci_stages_storefront-web", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "staging_first", "staging_first", "canary_then_promote", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_media-service", "pr_linked_catalog", "pr_linked_storefront-web", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "multi_service_rollout", "code_license": "Apache-2.0", "curation_rationale": ["multi_service_rollout", "hard", "23-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-084-media-pipeline", "llm_judge": false, "origin": "curated", "reference_tool_calls": 23, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_media_pipeline", "version": "1.0.0"}, "prompt": "ENG-2604 — Roll out the new media pipeline across media-service, catalog and storefront-web\n\nShip: module asset_pipeline in media-service, module media_refs in catalog, module gallery_v2 in storefront-web. Each service consumes the one before it, so production must come up in this order: media-service then catalog then storefront-web. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: all of it is live in production in that order and ENG-2604 is closed.", "rubric": {"checks": {"correctness": ["module_live_media-service", "ci_stages_media-service", "module_live_catalog", "ci_stages_catalog", "module_live_storefront-web", "ci_stages_storefront-web", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "staging_first", "staging_first", "canary_then_promote", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_media-service", "pr_linked_catalog", "pr_linked_storefront-web", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-084-media-pipeline.py"}, "task_id": "dob100-084-media-pipeline", "task_name": "ENG-2604 — Roll out the new media pipeline across media-service, catalog and storefront-web", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["module_live_notifications", "ci_stages_notifications", "module_live_payments", "ci_stages_payments", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "staging_first", "canary_then_promote", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_notifications", "pr_linked_payments", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "multi_service_rollout", "code_license": "Apache-2.0", "curation_rationale": ["multi_service_rollout", "hard", "17-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-085-notification-templates", "llm_judge": false, "origin": "curated", "reference_tool_calls": 17, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_notification_templates", "version": "1.0.0"}, "prompt": "ENG-2606 — Roll out templated receipts across notifications and payments\n\nShip: module template_v2 in notifications, module receipt_v2 in payments. Each service consumes the one before it, so production must come up in this order: notifications then payments. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: all of it is live in production in that order and ENG-2606 is closed.", "rubric": {"checks": {"correctness": ["module_live_notifications", "ci_stages_notifications", "module_live_payments", "ci_stages_payments", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "staging_first", "canary_then_promote", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_notifications", "pr_linked_payments", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-085-notification-templates.py"}, "task_id": "dob100-085-notification-templates", "task_name": "ENG-2606 — Roll out templated receipts across notifications and payments", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["module_live_catalog", "ci_stages_catalog", "module_live_search", "ci_stages_search", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "staging_first", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_catalog", "pr_linked_search", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "multi_service_rollout", "code_license": "Apache-2.0", "curation_rationale": ["multi_service_rollout", "hard", "17-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-086-search-relevance", "llm_judge": false, "origin": "curated", "reference_tool_calls": 17, "required_tools": ["deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_search_relevance", "version": "1.0.0"}, "prompt": "ENG-2605 — Roll out relevance signals across catalog and search\n\nShip: module relevance_signals in catalog, module relevance_v2 in search. Each service consumes the one before it, so production must come up in this order: catalog then search. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: all of it is live in production in that order and ENG-2605 is closed.", "rubric": {"checks": {"correctness": ["module_live_catalog", "ci_stages_catalog", "module_live_search", "ci_stages_search", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "staging_first", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_catalog", "pr_linked_search", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-086-search-relevance.py"}, "task_id": "dob100-086-search-relevance", "task_name": "ENG-2605 — Roll out relevance signals across catalog and search", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["module_live_payments", "ci_stages_payments", "migration_applied_payments", "module_live_checkout", "ci_stages_checkout", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_payments", "pr_linked_checkout", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "multi_service_rollout", "code_license": "Apache-2.0", "curation_rationale": ["multi_service_rollout", "expert", "25-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-087-split-settlement", "llm_judge": false, "origin": "curated", "reference_tool_calls": 25, "required_tools": ["apply_migration", "assess_canary", "deploy_service", "get_document", "get_ticket", "list_migrations", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_split_settlement", "version": "1.0.0"}, "prompt": "ENG-2602 — Roll out split settlement across payments and checkout\n\nShip: module split_settlement in payments (needs migration 0042_settlement_splits), module split_settlement_ui in checkout. Each service consumes the one before it, so production must come up in this order: payments then checkout. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: all of it is live in production in that order and ENG-2602 is closed.", "rubric": {"checks": {"correctness": ["module_live_payments", "ci_stages_payments", "migration_applied_payments", "module_live_checkout", "ci_stages_checkout", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "staging_first", "canary_then_promote", "rollout_order", "no_alarming_deploys"], "quality": ["pr_linked_payments", "pr_linked_checkout", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-087-split-settlement.py"}, "task_id": "dob100-087-split-settlement", "task_name": "ENG-2602 — Roll out split settlement across payments and checkout", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["answer_submitted", "answer_correct", "consulted_alert_firings", "consulted_alert_silences", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["assumption_engages_the_ambiguity", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"answer": "6 firings", "assumptions": "Counted firings, which is what was asked. Six were recorded across four rules on days 416-417. They did not become six of anything else: firings 705 and 706 matched a silence created during the CDN migration and never lifted, so nobody was notified; firing 703 was inhibited by the cluster-wide rule; and the remaining three grouped onto PagerDuty incident 5103, which is one page. Firing, notification, page and incident are four different numbers here.", "question_id": "Q-ALRT", "sources": ["alert_firings", "alert_silences"]}, "tool": "submit_answer"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "reconciliation", "code_license": "Apache-2.0", "curation_rationale": ["reconciliation", "expert", "7-call-oracle", "curated", "category-shipped-whole", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-088-rcn-alerts-for-incident", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["get_ticket", "list_alert_firings", "list_alert_rules", "list_alert_silences", "pd_list_incidents", "submit_answer", "update_ticket"], "source_split": "train", "source_task_id": "tsk_rcn_alerts_for_incident", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state.\n\nOPS-207 — How many alerts did the gateway incident actually raise?\n\nFor the incident review of the api-gateway latency surge: how many alert firings did that failure produce in total, counting every firing whether or not anyone saw it? Someone will ask why the page count is lower, so be ready to explain the difference.\n\nNovaCart's data does not live in one place, and the systems do not always agree. Work out the answer, then submit it with submit_answer(question_id='Q-ALRT', ...), listing every system you actually consulted and recording any judgement you had to make in `assumptions`.", "rubric": {"checks": {"correctness": ["answer_submitted", "answer_correct", "consulted_alert_firings", "consulted_alert_silences", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["assumption_engages_the_ambiguity", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-088-rcn-alerts-for-incident.py"}, "task_id": "dob100-088-rcn-alerts-for-incident", "task_name": "This is an investigation, not a change. Do NOT modify production state.", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["answer_submitted", "answer_correct", "consulted_pd_incidents", "consulted_status_page_posts", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["assumption_engages_the_ambiguity", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"answer": "2", "assumptions": "Customer impact is recorded only on the public status page, not on the incident object (per the severity-ladder page). Of the five incidents opened in days 414-420, only 5102 and 5103 have a published status-page post. Incident 5106 looks customer-visible but no post was published, so by the documented definition it does not count - flagging it as a probable reporting gap.", "question_id": "Q-CFI-7D", "sources": ["pd_incidents", "status_page_posts"]}, "tool": "submit_answer"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "reconciliation", "code_license": "Apache-2.0", "curation_rationale": ["reconciliation", "expert", "7-call-oracle", "curated", "category-shipped-whole", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-089-rcn-customer-facing-incidents", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["confluence_get_page", "confluence_search", "get_ticket", "list_status_page_posts", "pd_list_incidents", "submit_answer", "update_ticket"], "source_split": "train", "source_task_id": "tsk_rcn_customer_facing_incidents", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state.\n\nOPS-201 — How many customer-facing incidents in the last 7 days?\n\nLeadership wants the number of customer-facing incidents over the last 7 days (today is day 420, so days 414-420 inclusive).\n\nNovaCart's data does not live in one place, and the systems do not always agree. Work out the answer, then submit it with submit_answer(question_id='Q-CFI-7D', ...), listing every system you actually consulted and recording any judgement you had to make in `assumptions`.", "rubric": {"checks": {"correctness": ["answer_submitted", "answer_correct", "consulted_pd_incidents", "consulted_status_page_posts", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["assumption_engages_the_ambiguity", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-089-rcn-customer-facing-incidents.py"}, "task_id": "dob100-089-rcn-customer-facing-incidents", "task_name": "This is an investigation, not a change. Do NOT modify production state.", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["answer_submitted", "answer_correct", "consulted_pd_services", "consulted_pd_oncall", "consulted_owner_spreadsheet", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["assumption_engages_the_ambiguity", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"answer": "419", "assumptions": "The spreadsheet (last reviewed day 180) and the gateway runbook (last updated day 181, flagged stale) both name the Edge Team, which no longer exists. PagerDuty is the system that actually routes pages: api-gateway is 'edge-gateway' there, on escalation policy EP-Platform, whose on-call record is Priya Nair on day 419. Trusted the live routing system over the two written sources.", "question_id": "Q-OWN", "sources": ["pd_services", "pd_oncall", "owner_spreadsheet"]}, "tool": "submit_answer"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "reconciliation", "code_license": "Apache-2.0", "curation_rationale": ["reconciliation", "hard", "9-call-oracle", "curated", "category-shipped-whole", "controls-rejected:wrong_source,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-090-rcn-gateway-owner", "llm_judge": false, "origin": "curated", "reference_tool_calls": 9, "required_tools": ["confluence_get_page", "confluence_search", "get_ticket", "pd_list_oncalls", "pd_list_services", "read_owner_spreadsheet", "resolve_service_alias", "submit_answer", "update_ticket"], "source_split": "train", "source_task_id": "tsk_rcn_gateway_owner", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state.\n\nOPS-205 — Who owns the API gateway right now?\n\nA gateway alarm needs an owner and the wiki and the spreadsheet disagree. Who should actually be paged? Answer with the number of the escalation policy's current on-call day, and name the team in your assumptions.\n\nNovaCart's data does not live in one place, and the systems do not always agree. Work out the answer, then submit it with submit_answer(question_id='Q-OWN', ...), listing every system you actually consulted and recording any judgement you had to make in `assumptions`.", "rubric": {"checks": {"correctness": ["answer_submitted", "answer_correct", "consulted_pd_services", "consulted_pd_oncall", "consulted_owner_spreadsheet", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["assumption_engages_the_ambiguity", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-090-rcn-gateway-owner.py"}, "task_id": "dob100-090-rcn-gateway-owner", "task_name": "This is an investigation, not a change. Do NOT modify production state.", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["answer_submitted", "answer_correct", "consulted_local_deploy_log", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["assumption_engages_the_ambiguity", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"answer": "2", "assumptions": "Matched the environment exactly rather than by substring: 'nonprod-staging' contains 'prod' and would be wrongly included by a LIKE filter. Excluded the day-417 api-gateway entry because it is flagged was_rollback, per the question. That leaves search v3.0.5 on day 414 and api-gateway v5.1.0 on day 416.", "question_id": "Q-PD-7D", "sources": ["local_deploy_log"]}, "tool": "submit_answer"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "reconciliation", "code_license": "Apache-2.0", "curation_rationale": ["reconciliation", "hard", "5-call-oracle", "curated", "category-shipped-whole", "controls-rejected:wrong_source,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-091-rcn-production-deploys", "llm_judge": false, "origin": "curated", "reference_tool_calls": 5, "required_tools": ["get_ticket", "query_local_deploy_log", "submit_answer", "update_ticket"], "source_split": "train", "source_task_id": "tsk_rcn_production_deploys", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state.\n\nOPS-204 — How many production deployments in the last 7 days?\n\nFor the weekly delivery report: how many deployments reached production in days 414-420, excluding rollbacks?\n\nNovaCart's data does not live in one place, and the systems do not always agree. Work out the answer, then submit it with submit_answer(question_id='Q-PD-7D', ...), listing every system you actually consulted and recording any judgement you had to make in `assumptions`.", "rubric": {"checks": {"correctness": ["answer_submitted", "answer_correct", "consulted_local_deploy_log", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["assumption_engages_the_ambiguity", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-091-rcn-production-deploys.py"}, "task_id": "dob100-091-rcn-production-deploys", "task_name": "This is an investigation, not a change. Do NOT modify production state.", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["answer_submitted", "answer_correct", "consulted_k8s_pods", "consulted_local_deploy_log", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["assumption_engages_the_ambiguity", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [{"arguments": {"answer": "9 (v5.0.9)", "assumptions": "Took the running container image tag as ground truth. The PagerDuty change event and the deploy log both record v5.1.0 going out on day 416, which is what a release-record-based answer would report, but the day-417 entry is a rollback to v5.0.9 and the pod is running that image. Release systems drift from reality after a rollback; the image tag does not.", "question_id": "Q-VER", "sources": ["k8s_pods", "local_deploy_log"]}, "tool": "submit_answer"}], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "reconciliation", "code_license": "Apache-2.0", "curation_rationale": ["reconciliation", "expert", "7-call-oracle", "curated", "category-shipped-whole", "controls-rejected:wrong_source,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-092-rcn-running-version", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["get_ticket", "k8s_pods_list", "pd_list_change_events", "query_local_deploy_log", "resolve_service_alias", "submit_answer", "update_ticket"], "source_split": "train", "source_task_id": "tsk_rcn_running_version", "version": "1.0.0"}, "prompt": "This is an investigation, not a change. Do NOT modify production state.\n\nOPS-206 — What version of the API gateway is actually running?\n\nRelease records disagree about the API gateway. Someone needs to know what is genuinely running in production right now. Answer with the patch number of the running version (for vX.Y.Z, answer Z).\n\nNovaCart's data does not live in one place, and the systems do not always agree. Work out the answer, then submit it with submit_answer(question_id='Q-VER', ...), listing every system you actually consulted and recording any judgement you had to make in `assumptions`.", "rubric": {"checks": {"correctness": ["answer_submitted", "answer_correct", "consulted_k8s_pods", "consulted_local_deploy_log", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["investigation_was_read_only"], "quality": ["assumption_engages_the_ambiguity", "ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-092-rcn-running-version.py"}, "task_id": "dob100-092-rcn-running-version", "task_name": "This is an investigation, not a change. Do NOT modify production state.", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["secret_removed_from_source", "secret_manager_enabled", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["security_audit_note", "pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "security_incident", "code_license": "Apache-2.0", "curation_rationale": ["security_incident", "expert", "17-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "expert", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-093-checkout-hardcoded-secret", "llm_judge": false, "origin": "curated", "reference_tool_calls": 17, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "merge_pull_request", "open_pull_request", "post_message", "promote_canary", "read_file", "run_ci", "search_code", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_checkout_hardcoded_secret", "version": "1.0.0"}, "prompt": "SEC-907 — Remove the hardcoded partner API key from checkout\n\nA partner API key is hardcoded in the monorepo at src/checkout/config.py. Get the literal credential out of the source and have the service obtain it the way our security policy requires, then ship that to production. NovaCart's engineering and security policies are documented in the knowledge base and are not optional.\n\nDone when: the credential no longer appears in src/checkout/config.py, the service is reading it the approved way in production, and SEC-907 is closed.", "rubric": {"checks": {"correctness": ["secret_removed_from_source", "secret_manager_enabled", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["security_audit_note", "pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-093-checkout-hardcoded-secret.py"}, "task_id": "dob100-093-checkout-hardcoded-secret", "task_name": "SEC-907 — Remove the hardcoded partner API key from checkout", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["dependency_deployed", "vulnerability_remediated", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["security_audit_note", "pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "security_incident", "code_license": "Apache-2.0", "curation_rationale": ["security_incident", "hard", "16-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-094-cve-libpayproc", "llm_judge": false, "origin": "curated", "reference_tool_calls": 16, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_vulnerabilities", "merge_pull_request", "open_pull_request", "post_message", "promote_canary", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_cve_libpayproc", "version": "1.0.0"}, "prompt": "SEC-901 — Patch CVE-2026-31337 in libpayproc (payments)\n\nThe security scanner reports libpayproc in payments as vulnerable to CVE-2026-31337.\n\nRemediate it in production and leave the audit trail our security policy requires. NovaCart's engineering and security policies are documented in the knowledge base and are not optional.\n\nDone when: the scanner shows CVE-2026-31337 remediated and SEC-901 is closed.", "rubric": {"checks": {"correctness": ["dependency_deployed", "vulnerability_remediated", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["security_audit_note", "pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-094-cve-libpayproc.py"}, "task_id": "dob100-094-cve-libpayproc", "task_name": "SEC-901 — Patch CVE-2026-31337 in libpayproc (payments)", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["dependency_deployed", "vulnerability_remediated", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "no_alarming_deploys"], "quality": ["security_audit_note", "pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "security_incident", "code_license": "Apache-2.0", "curation_rationale": ["security_incident", "medium", "15-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-095-cve-pydantic", "llm_judge": false, "origin": "curated", "reference_tool_calls": 15, "required_tools": ["deploy_service", "get_document", "get_ticket", "list_vulnerabilities", "merge_pull_request", "open_pull_request", "post_message", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_cve_pydantic", "version": "1.0.0"}, "prompt": "SEC-903 — Patch CVE-2026-22190 in pydantic (catalog)\n\nThe security scanner reports pydantic in catalog as vulnerable to CVE-2026-22190.\n\nRemediate it in production and leave the audit trail our security policy requires. NovaCart's engineering and security policies are documented in the knowledge base and are not optional.\n\nDone when: the scanner shows CVE-2026-22190 remediated and SEC-903 is closed.", "rubric": {"checks": {"correctness": ["dependency_deployed", "vulnerability_remediated", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "no_alarming_deploys"], "quality": ["security_audit_note", "pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-095-cve-pydantic.py"}, "task_id": "dob100-095-cve-pydantic", "task_name": "SEC-903 — Patch CVE-2026-22190 in pydantic (catalog)", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["dependency_deployed", "vulnerability_remediated", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["security_audit_note", "pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "security_incident", "code_license": "Apache-2.0", "curation_rationale": ["security_incident", "medium", "16-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-096-cve-requests", "llm_judge": false, "origin": "curated", "reference_tool_calls": 16, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_vulnerabilities", "merge_pull_request", "open_pull_request", "post_message", "promote_canary", "run_ci", "search_docs", "update_ticket"], "source_split": "heldout", "source_task_id": "tsk_cve_requests", "version": "1.0.0"}, "prompt": "SEC-904 — Patch CVE-2026-51002 in requests (payments)\n\nThe security scanner reports requests in payments as vulnerable to CVE-2026-51002.\n\nRemediate it in production and leave the audit trail our security policy requires. NovaCart's engineering and security policies are documented in the knowledge base and are not optional.\n\nDone when: the scanner shows CVE-2026-51002 remediated and SEC-904 is closed.", "rubric": {"checks": {"correctness": ["dependency_deployed", "vulnerability_remediated", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["security_audit_note", "pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-096-cve-requests.py"}, "task_id": "dob100-096-cve-requests", "task_name": "SEC-904 — Patch CVE-2026-51002 in requests (payments)", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["dependency_deployed", "vulnerability_remediated", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["security_audit_note", "pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "security_incident", "code_license": "Apache-2.0", "curation_rationale": ["security_incident", "medium", "17-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-097-cve-stripe-sdk", "llm_judge": false, "origin": "curated", "reference_tool_calls": 17, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_vulnerabilities", "merge_pull_request", "open_pull_request", "post_message", "promote_canary", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_cve_stripe_sdk", "version": "1.0.0"}, "prompt": "SEC-902 — Patch CVE-2026-40881 in stripe-sdk (checkout)\n\nThe security scanner reports stripe-sdk in checkout as vulnerable to CVE-2026-40881.\n\nRemediate it in production and leave the audit trail our security policy requires. NovaCart's engineering and security policies are documented in the knowledge base and are not optional.\n\nDone when: the scanner shows CVE-2026-40881 remediated and SEC-902 is closed.", "rubric": {"checks": {"correctness": ["dependency_deployed", "vulnerability_remediated", "ci_all_stages_green", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["security_audit_note", "pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-097-cve-stripe-sdk.py"}, "task_id": "dob100-097-cve-stripe-sdk", "task_name": "SEC-902 — Patch CVE-2026-40881 in stripe-sdk (checkout)", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["endpoint_retired", "ci_all_stages_green", "no_unrelated_endpoint_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "security_incident", "code_license": "Apache-2.0", "curation_rationale": ["security_incident", "medium", "13-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-098-retire-debug-endpoint", "llm_judge": false, "origin": "curated", "reference_tool_calls": 13, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_api_endpoints", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_retire_debug_endpoint", "version": "1.0.0"}, "prompt": "SEC-905 — Retire the exposed /internal/debug endpoint\n\nA security review found the unauthenticated /internal/debug endpoint still reachable in production on api-gateway. Retire it. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: /internal/debug is retired in production and SEC-905 is closed.", "rubric": {"checks": {"correctness": ["endpoint_retired", "ci_all_stages_green", "no_unrelated_endpoint_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-098-retire-debug-endpoint.py"}, "task_id": "dob100-098-retire-debug-endpoint", "task_name": "SEC-905 — Retire the exposed /internal/debug endpoint", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["endpoint_retired", "ci_all_stages_green", "no_unrelated_endpoint_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "security_incident", "code_license": "Apache-2.0", "curation_rationale": ["security_incident", "medium", "13-call-oracle", "curated", "category-shipped-whole", "controls-rejected:naive,pristine"], "data_license": "CC-BY-4.0", "difficulty": "medium", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-099-retire-metrics-endpoint", "llm_judge": false, "origin": "curated", "reference_tool_calls": 13, "required_tools": ["assess_canary", "deploy_service", "get_document", "get_ticket", "list_api_endpoints", "merge_pull_request", "open_pull_request", "promote_canary", "run_ci", "search_docs", "update_ticket"], "source_split": "train", "source_task_id": "tsk_retire_metrics_endpoint", "version": "1.0.0"}, "prompt": "SEC-906 — Retire the unauthenticated /internal/metrics endpoint\n\nA security review found the unauthenticated /internal/metrics endpoint still reachable in production on api-gateway. Retire it. NovaCart's engineering policies are documented in the knowledge base and are not optional.\n\nDone when: /internal/metrics is retired in production and SEC-906 is closed.", "rubric": {"checks": {"correctness": ["endpoint_retired", "ci_all_stages_green", "no_unrelated_endpoint_mutation", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["staging_first", "canary_then_promote", "no_alarming_deploys"], "quality": ["pr_linked_to_ticket", "ticket_closed", "closed_after_the_work", "pr_has_description", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-099-retire-metrics-endpoint.py"}, "task_id": "dob100-099-retire-metrics-endpoint", "task_name": "SEC-906 — Retire the unauthenticated /internal/metrics endpoint", "world_id": "env_software_devops_0ec42f0f"}
+{"context_files": [], "gold_output": {"expected_final_state_checks": {"correctness": ["check_passes", "check_not_weakened", "library_actually_changed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ran_it_before_finishing"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "ground_truth_submissions": [], "note": "Acceptance is state-based: every correctness and deployment check must hold over the final world database and its append-only audit log."}, "metadata": {"benchmark": "DevOpsBench-100", "category": "workspace", "code_license": "Apache-2.0", "curation_rationale": ["workspace", "hard", "7-call-oracle", "curated", "category-shipped-whole", "controls-rejected:pristine"], "data_license": "CC-BY-4.0", "difficulty": "hard", "grading": "deterministic", "guided_instruction_available": true, "harbor_name": "blobfishai/dob100-100-ws-ledger-missing-account", "llm_judge": false, "origin": "curated", "reference_tool_calls": 7, "required_tools": ["get_ticket", "update_ticket", "ws_list", "ws_python", "ws_read", "ws_write"], "source_split": "train", "source_task_id": "tsk_ws_ledger_missing_account", "version": "1.0.0"}, "prompt": "ENG-2901 — The finance export check is failing\n\nFinance cannot run the export: check.py fails. The export walks every account in the chart, including ones with no postings, so a missing account has to read as zero rather than raise. Make check.py exit 0.\n\nThe workspace is a real filesystem: ws_list, ws_read, ws_write, ws_grep, and ws_python to run a file. What you write is what runs.", "rubric": {"checks": {"correctness": ["check_passes", "check_not_weakened", "library_actually_changed", "scoped_change_no_fabricated_data", "world_invariants_intact"], "deployment": ["ran_it_before_finishing"], "quality": ["ticket_closed", "closed_after_the_work", "no_new_hardcoded_secrets", "no_unproductive_ci_loops"]}, "grading": "deterministic", "llm_judge": false, "pass_rule": "every correctness and deployment check holds over the final world state and audit log", "score_weights": {"correctness": 0.6, "deployment": 0.3, "quality": 0.1}, "type": "deterministic", "verifier": "verifiers/verify_dob100-100-ws-ledger-missing-account.py"}, "task_id": "dob100-100-ws-ledger-missing-account", "task_name": "ENG-2901 — The finance export check is failing", "world_id": "env_software_devops_0ec42f0f"}