fix(governance): trace enum typos, surface review-due badge, retain safety-citation provenance (audit #1/#2/#9/#13)#1051
Merged
Conversation
…afety-citation provenance Resolves the regression-safe subset of the 13-issue source-governance audit: - #1 source-metadata: logger.warn on a present-but-unrecognized document_status / clinical_validation_status / extraction_quality value (data-entry typo) instead of silently collapsing it into the fallback. Return value unchanged — observability only. - #2 source-governance: add review_due_source to the frontend-visible warning set so a source past its review date shows a governance badge, consistent with the review-due language already emitted in the render-policy copy text. Warning severity — never forces the danger refusal. - #9 answer-response: retain source_metadata on safety-finding citations at the browser boundary (regular source citations already keep it), and render a governance pill in the safety-findings panel for outdated / review-due / unverified sources. - #13 tests: new provenance-flow.test.ts asserting governance metadata survives DB-normalization -> governance warnings -> governed client payload (sources AND safety citations) -> render policy — the end-to-end guard that would have caught #9. RAG impact: no retrieval behaviour change — governance-metadata display/observability only; no files on the ranking/eval/selection surface (verified against ragRankingPatterns). Deferred (need a live eval-canary + approval, or are architectural): audit #4/#5/#6/#8/#10 (RAG-protected ranking/generation/cache), #11/#5 flag debt, #3 (is_public schema), #7 (conflict-detection scope), #12 (canary automation). Verification: verify:cheap (340 files / 3068 tests), typecheck, lint, prettier all green. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
|
This pull request has been ignored for the connected project Preview Branches by Supabase. |
Contributor
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (2)
🚧 Files skipped from review as they are similar to previous changes (1)
📝 WalkthroughWalkthroughGovernance metadata now logs invalid enum values, remains attached to governed safety citations, exposes review-due warnings, and appears as warning badges on clinical safety findings. Tests cover normalization, end-to-end propagation, frontend visibility, and refusal for outdated supporting sources. ChangesGovernance provenance flow
Estimated code review effort: 3 (Moderate) | ~25 minutes Sequence Diagram(s)sequenceDiagram
participant Source as Source metadata
participant Normalizer as normalizeSourceMetadata
participant Governance as frontendSourceGovernanceWarnings
participant Payload as clientSafetyWarning
participant Dashboard as SafetyFindingsListContent
Source->>Normalizer: provide document and validation statuses
Normalizer->>Governance: return normalized metadata and warnings
Governance->>Payload: expose review_due and validation warnings
Payload->>Dashboard: retain citation source_metadata
Dashboard->>Dashboard: render governance warning badges
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@tests/source-metadata.test.ts`:
- Around line 23-36: Extend the test case around normalizeSourceMetadata to spy
on logger.warn and assert it is called for each of the three invalid non-empty
enum values. Add coverage for absent and blank inputs, verifying they do not
trigger warnings while preserving the existing fallback assertions.
🪄 Autofix (Beta)
❌ Autofix failed (check again to retry)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro Plus
Run ID: 9cf0602b-4b82-4c9f-9420-147336e85e4f
📒 Files selected for processing (8)
src/components/clinical-dashboard/evidence-panels.tsxsrc/lib/answer-response.tssrc/lib/source-governance.tssrc/lib/source-metadata.tstests/answer-client-payload.test.tstests/provenance-flow.test.tstests/source-governance.test.tstests/source-metadata.test.ts
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@docs/branch-review-ledger.md`:
- Line 684: Update the ledger entry for claude/database-governance-audit-10b6ed
(PR `#1051`) to use the actual current review/merge date, 2026-07-21, instead of
the future date 2026-07-22. Preserve the rest of the entry unchanged.
🪄 Autofix (Beta)
✅ Autofix completed
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro Plus
Run ID: 2cee8004-9cbd-45da-8828-69153620893c
📒 Files selected for processing (1)
docs/branch-review-ledger.md
BigSimmo
enabled auto-merge (squash)
July 22, 2026 01:19
Contributor
|
Note Autofix is a beta feature. Expect some limitations and changes as we gather feedback and continue to improve it. The branch was updated while autofix was in progress. Please try again. |
Contributor
|
Note Autofix is a beta feature. Expect some limitations and changes as we gather feedback and continue to improve it. Fixes Applied SuccessfullyFixed 1 file(s) based on 1 unresolved review comment. Files modified:
Commit: The changes have been pushed to the Time taken: |
Fixed 1 file(s) based on 1 unresolved review comment. Co-authored-by: CodeRabbit <noreply@coderabbit.ai>
…nance-audit-10b6ed
…ce-metadata enums Address the CodeRabbit finding on the issue-#1 enum test: spy on logger.warn and assert it fires once per present-but-unrecognized document_status / clinical_validation_status / extraction_quality value (with field + value), while absent/blank inputs stay silent. Keeps the existing safe-fallback assertions — the return value is unchanged, so this pins the observability behaviour without altering any downstream ranking/rendering. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LRZ1EyBZW1ADXrZvMxEEsC
…nance-audit-10b6ed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Resolves the regression-safe subset of the 13-issue source-governance audit. The ranking/generation issues were deferred because they sit on the RAG-behaviour-protected surface and need a live eval-canary + approval (see below).
enumOrDefault(src/lib/source-metadata.ts) nowlogger.warns a present-but-unrecognizeddocument_status/clinical_validation_status/extraction_qualityvalue (a data-entry typo) instead of silently collapsing it into the fallback. Absent/empty stays silent. Return value unchanged — observability only.review_due_sourcetofrontendVisibleWarningCodes(src/lib/source-governance.ts) so a source past its review date shows a governance badge. Warning-severity → never forces the danger refusal. Also reconciles a real inconsistency where the render-policy copy-text already surfaced review-due language while the badge list suppressed it.clientSafetyWarning(src/lib/answer-response.ts) now retainssource_metadataon safety-finding citations (regular source citations already keep it), and the safety-findings panel (evidence-panels.tsx) renders a governance pill for outdated / review-due / unverified sources.tests/provenance-flow.test.tsasserting governance metadata survives DB-normalization → governance warnings → governed client payload (sources and safety citations) → render policy. This is the guard that would have caught Bump lucide-react from 1.16.0 to 1.17.0 #9.RAG impact: no retrieval behaviour change — governance-metadata display/observability only; no files on the ranking/eval/selection surface (verified against
ragRankingPatternsinscripts/pr-policy.mjs;check:pr-policygreen).Deferred (need a live eval-canary + approval, or architectural): #4/#5/#6/#8/#10 (RAG-protected ranking / best-source selection / LLM-context / cache), #11 & #5 (D5/D4 flag debt), #3 (
is_publicschema/RLS), #7 (conflict-detection scope), #12 (canary automation).Verification
npm run verify:pr-local— exit 0: 351 files / 3129 tests, production build, client-bundle secret scan, offline RAG fixtures (36 golden cases) all green.globals.css(unmodified by this PR;git-clean; the production build insideverify:pr-localpassed, so it is dev-Turbopack-only). The UI change is an additive, conditionally-rendered governance pill using existing primitives, covered by typecheck + lint + unit tests.eval:retrieval:qualitynot required.check:production-readinessnot run locally: it is an offline env/config validator and this PR changes no env/secret/config inputs; the worktree is secretless (demo mode) so it would only report expected missing-secret state. It runs in CI.Risk and rollout
Clinical Governance Preflight
Clinical KB Database(sjrfecxgysukkwxsowpy) — no env/config touchedoutdatedstill triggers the danger refusal, unchangedNotes
origin/main(was 18 behind); rebase was conflict-free (none of the 18 commits touched the 8 changed files).🤖 Generated with Claude Code
Summary by CodeRabbit