vignette/apps/api/app/data/supervision_research_benchmark_g6.v1.json
Yun Chan 16e791e044 G0~G8 성과·동맹 측정 OS 작업 일괄 고정
8월 7일까지 워킹트리에만 남아 있던 미커밋 작업을 커밋한다. 여러 사본
폴더(worktree·clone)에 흩어져 있던 중간 스냅샷을 정리하기 전에 원본을
git 이력으로 고정하는 것이 목적이다.

- contracts/routes/services: measurement, outcome_trajectory, rupture_repair,
  deliberate_practice, calibration_transfer, supervision_research,
  multimodal_alliance, continuous_improvement 계열 신규 모듈과 테스트
- infra/db/init: 07~16 마이그레이션(측정 기반~calibration transfer 실행)
- apps/web: 세션 리뷰 카드·관리 화면·E2E 스펙 추가
- docs/ops: G0~G8 라이브 통합·배포·롤백 증거 문서와 evidence JSON/PNG
- scripts: smoke·ledger·릴리스 에이전트·NAS 프리뷰 운영 스크립트

engine.public 로그 .bak과 apps/web/test-results 산출물은 커밋에서 제외했다.
2026-08-08 01:30:53 +09:00

47 lines
6.8 KiB
JSON

{
"schema_version": "vignette.supervision-research-benchmark.v1",
"version": "1.0.0",
"data_classification": "synthetic_educational",
"clinical_claim_allowed": false,
"attention_signals": [
{"signal_id":"oas-g6-signal-deterioration","learner_ref":"learner-d","signal_type":"deterioration","severity":"high","state":"active","uncertainty":0.1,"observed_sequence":6,"evidence":[{"ledger":"outcome_trajectory_revision","event_id":"trajectory-d","session_id":"session-d","route_hint":"/teach/analysis/learner-d?tab=outcome"}],"counterevidence":[]},
{"signal_id":"oas-g6-signal-rupture","learner_ref":"learner-r","signal_type":"unresolved_rupture","severity":"high","state":"active","uncertainty":0.2,"observed_sequence":5,"evidence":[{"ledger":"rupture_observation_event","event_id":"rupture-r","session_id":"session-r","route_hint":"/teach/analysis/learner-r?tab=rupture"}],"counterevidence":[]},
{"signal_id":"oas-g6-signal-safety","learner_ref":"learner-s","signal_type":"safety_boundary","severity":"high","state":"active","uncertainty":0.0,"observed_sequence":4,"evidence":[{"ledger":"safety_event","event_id":"safety-s","session_id":"session-s","route_hint":"/teach/analysis/learner-s?tab=safety"}],"counterevidence":[]},
{"signal_id":"oas-g6-signal-calibration","learner_ref":"learner-c","signal_type":"persistent_overconfidence","severity":"moderate","state":"monitoring","uncertainty":0.2,"observed_sequence":3,"evidence":[{"ledger":"calibration_assessment","event_id":"calibration-c","session_id":null,"route_hint":"/teach/analysis/learner-c?tab=calibration"}],"counterevidence":[]},
{"signal_id":"oas-g6-signal-stagnation","learner_ref":"learner-g","signal_type":"growth_stagnation","severity":"moderate","state":"active","uncertainty":0.3,"observed_sequence":2,"evidence":[{"ledger":"practice_attempt","event_id":"practice-g","session_id":"session-g","route_hint":"/teach/analysis/learner-g?tab=practice"}],"counterevidence":[]},
{"signal_id":"oas-g6-signal-transfer","learner_ref":"learner-t","signal_type":"transfer_failure","severity":"moderate","state":"active","uncertainty":0.2,"observed_sequence":1,"evidence":[{"ledger":"transfer_assessment","event_id":"transfer-t","session_id":null,"route_hint":"/teach/analysis/learner-t?tab=transfer"}],"counterevidence":[]}
],
"expected_queue_order": ["learner-s","learner-d","learner-r","learner-c","learner-g","learner-t"],
"disagreements": [
{"disagreement_id":"oas-g6-disagreement-001","case_ref":"synthetic-case-001","competency_id":"competency.rupture-repair","ai_label":"resolved","teacher_label":"partial","ai_model":"evaluator-v1","prompt_version":"1.0.0","instrument_id":"rupture-repair-evaluator","instrument_version":"1.0.0","ai_evidence":[{"ledger":"rupture_reconciliation_revision","event_id":"reconcile-ai-001","session_id":"session-001","route_hint":"/sessions/session-001/ruptures"}],"teacher_correction_evidence":[{"ledger":"rupture_observation_event","event_id":"correction-human-001","session_id":"session-001","route_hint":"/sessions/session-001/ruptures"}],"correction_reason_code":"repair_impact_not_confirmed"}
],
"baseline_batch": {
"batch_id":"oas-g6-batch-baseline","model":"evaluator-v1","prompt_version":"1.0.0","instrument_id":"rupture-eval","instrument_version":"1.0.0",
"observations":[
{"case_ref":"case-a1","competency_id":"competency.repair","synthetic_subgroup":"synthetic-a","gold_label":"resolved","predicted_label":"resolved","evidence_event_id":"base-a1"},
{"case_ref":"case-a2","competency_id":"competency.repair","synthetic_subgroup":"synthetic-a","gold_label":"partial","predicted_label":"partial","evidence_event_id":"base-a2"},
{"case_ref":"case-a3","competency_id":"competency.repair","synthetic_subgroup":"synthetic-a","gold_label":"missed","predicted_label":"missed","evidence_event_id":"base-a3"},
{"case_ref":"case-b1","competency_id":"competency.repair","synthetic_subgroup":"synthetic-b","gold_label":"resolved","predicted_label":"resolved","evidence_event_id":"base-b1"},
{"case_ref":"case-b2","competency_id":"competency.repair","synthetic_subgroup":"synthetic-b","gold_label":"partial","predicted_label":"partial","evidence_event_id":"base-b2"},
{"case_ref":"case-b3","competency_id":"competency.repair","synthetic_subgroup":"synthetic-b","gold_label":"missed","predicted_label":"missed","evidence_event_id":"base-b3"}
]
},
"candidate_batch": {
"batch_id":"oas-g6-batch-candidate","model":"evaluator-v2","prompt_version":"2.0.0","instrument_id":"rupture-eval","instrument_version":"1.1.0",
"observations":[
{"case_ref":"case-a1","competency_id":"competency.repair","synthetic_subgroup":"synthetic-a","gold_label":"resolved","predicted_label":"resolved","evidence_event_id":"candidate-a1"},
{"case_ref":"case-a2","competency_id":"competency.repair","synthetic_subgroup":"synthetic-a","gold_label":"partial","predicted_label":"partial","evidence_event_id":"candidate-a2"},
{"case_ref":"case-a3","competency_id":"competency.repair","synthetic_subgroup":"synthetic-a","gold_label":"missed","predicted_label":"missed","evidence_event_id":"candidate-a3"},
{"case_ref":"case-b1","competency_id":"competency.repair","synthetic_subgroup":"synthetic-b","gold_label":"resolved","predicted_label":"resolved","evidence_event_id":"candidate-b1"},
{"case_ref":"case-b2","competency_id":"competency.repair","synthetic_subgroup":"synthetic-b","gold_label":"partial","predicted_label":"resolved","evidence_event_id":"candidate-b2"},
{"case_ref":"case-b3","competency_id":"competency.repair","synthetic_subgroup":"synthetic-b","gold_label":"missed","predicted_label":"partial","evidence_event_id":"candidate-b3"}
]
},
"expected_drift_status": "drift_flagged",
"phase3_artifacts": [
{"domain":"alliance","artifact_id":"alliance-v1","schema_version":"vignette.alliance-pulse.v1","content_sha256":"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa","record_count":9,"provenance_uri":"db://app/measurement_event/alliance","clinical_claim_allowed":false},
{"domain":"rupture","artifact_id":"rupture-v1","schema_version":"vignette.rupture-repair.v1","content_sha256":"bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb","record_count":6,"provenance_uri":"db://app/rupture_observation_event","clinical_claim_allowed":false},
{"domain":"transfer","artifact_id":"transfer-v1","schema_version":"vignette.transfer-assessment.v1","content_sha256":"cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc","record_count":4,"provenance_uri":"audit://transfer/assessment","clinical_claim_allowed":false},
{"domain":"calibration","artifact_id":"calibration-v1","schema_version":"vignette.calibration-assessment.v1","content_sha256":"dddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd","record_count":6,"provenance_uri":"audit://calibration/assessment","clinical_claim_allowed":false}
]
}