Tracks the first formal independent clinical-review batch defined in docs/CLINICAL_REVIEW_PROGRAM.md.
External-review package:
docs/TIER_A_EXTERNAL_REVIEW_PACK.mddocs/TIER_A_EXTERNAL_REVIEW_HANDOFF.md- original AI-assisted audit:
docs/TIER_A_AI_CLINICAL_AUDIT_2026-09-19.md
- adaptive-flow re-audit addendum:
docs/TIER_A_ADAPTIVE_FLOW_REAUDIT_2026-09-19.md
- structured per-tool template:
.github/ISSUE_TEMPLATE/clinical-review.yml
Frozen clinical implementation baseline: a9d2bfcb569600a5e0c9be681b729cc2bf4ba661
The AI-assisted clinical-technical audit, remediation, dedicated boundary testing and adaptive-flow re-audit are complete. They do not count as independent clinical review.
Review each tool against the frozen clinical implementation SHA and authoritative sources. Documentation-only commits after that SHA do not alter the implementation under review. Do not include identifiable patient data.
For each completed review, record reviewer identity/role, date, commit/release, sources checked, ordinary and boundary fixtures, missing-data verification, ES/EN terminology review and one of the outcomes defined in the review program.
Any later logic-changing correction requires a dedicated PR and re-review of the affected implementation before that tool is marked independently reviewed.
Tracks the first formal independent clinical-review batch defined in
docs/CLINICAL_REVIEW_PROGRAM.md.External-review package:
docs/TIER_A_EXTERNAL_REVIEW_PACK.mddocs/TIER_A_EXTERNAL_REVIEW_HANDOFF.mddocs/TIER_A_AI_CLINICAL_AUDIT_2026-09-19.mddocs/TIER_A_ADAPTIVE_FLOW_REAUDIT_2026-09-19.md.github/ISSUE_TEMPLATE/clinical-review.ymlFrozen clinical implementation baseline:
a9d2bfcb569600a5e0c9be681b729cc2bf4ba661The AI-assisted clinical-technical audit, remediation, dedicated boundary testing and adaptive-flow re-audit are complete. They do not count as independent clinical review.
Review each tool against the frozen clinical implementation SHA and authoritative sources. Documentation-only commits after that SHA do not alter the implementation under review. Do not include identifiable patient data.
For each completed review, record reviewer identity/role, date, commit/release, sources checked, ordinary and boundary fixtures, missing-data verification, ES/EN terminology review and one of the outcomes defined in the review program.
Any later logic-changing correction requires a dedicated PR and re-review of the affected implementation before that tool is marked independently reviewed.