Claude/gate answer persistence flag 4b55da

[BigSimmo]

Summary

Verification

• npm run verify:pr-local

During development, use npm run verify:cheap as the faster iteration gate before the final PR-local preflight.

• npm run verify:ui when UI, routing, styling, browser behavior, reduced-motion, or forced-colors behavior changed
• npm run verify:release before release or handoff confidence claims

For retrieval, ranking, selection, chunking, source/citation rendering, or answer-contract changes, verify:pr-local runs eval:rag:offline automatically. Run the offline command directly during iteration before spending a live eval.

• npm run eval:retrieval:quality (must stay 36/36) when retrieval, ranking, selection, chunking, or scoring behavior changed — CI cannot run it (needs live keys), so run it locally and paste the summary. A metadata/governance-weighting change once buried correct docs (recall 1.0→0.76) and only this eval caught it.
• npm run eval:rag -- --limit 15 + npm run eval:quality -- --rag-only when answer generation, the synthesis prompt, or answer post-processing changed (grounded-supported must not drop; citation-failure 0)
• npm run check:production-readiness when clinical workflow, privacy, environment, Supabase, source governance, or deployment behavior changed
• npm run check:deployment-readiness when deployment startup, hosting, or rollout behavior changed

Clinical Governance Preflight

Complete this section when the change touches ingestion, answer generation, search/ranking, source rendering, document access, privacy, production env, or clinical output.

• Source-backed claims still require linked source verification before clinical use
• No patient-identifiable document workflow was introduced or expanded without explicit governance approval
• Supabase target remains Clinical KB Database (sjrfecxgysukkwxsowpy)
• Service-role keys and private document access remain server-only
• Demo/synthetic content remains clearly separated from real clinical sources
• Source metadata, review status, and outdated/unknown-source behavior remain conservative
• Deployment classification/TGA SaMD impact was checked when clinical decision-support behavior changed

Notes

[supabase]

This pull request has been ignored for the connected project sjrfecxgysukkwxsowpy because there are no changes detected in supabase directory. You can change this behaviour in Project Integrations Settings ↗︎.

---

Preview Branches by Supabase.
Learn more about Supabase Branching ↗︎.

[BigSimmo]

@copilot resolve the merge conflicts in this pull request

[coderabbitai]

Caution

Review failed

The head commit changed during the review from a3f498e to 47206d8.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch claude/gate-answer-persistence-flag-4b55da

✨ Simplify code

• Create PR with simplified code
• Commit simplified code in branch claude/gate-answer-persistence-flag-4b55da

---

_{Comment @coderabbitai help to get the list of available commands.}

[Copilot]

@copilot resolve the merge conflicts in this pull request

Done — resolved all merge conflicts in docs/privacy-impact-assessment.md (47206d8). The 5 conflict regions all arose from origin/main adding PIA-1/PIA-5 progress notes (PR #513), an updated rag_response_cache retention description, and a revised §11 recommendation; I took the origin/main side for each, which is the most current state.

[chatgpt-codex-connector]

💡 Codex Review

Database/src/lib/env.ts

Lines 116 to 120 in c26e54f

	// answer is not). Default OFF: do not persist generated answer text at rest.
	// The offline eval/quality pipeline reads the in-memory answer (logQuery:false)
	// and never reads this column back, so persistence-off is safe. Set true only
	// where retaining answer text is permitted and a retention policy exists
	// (owner-scoped + 30-day purge). Blocked in production readiness.

Gate cached answers behind the retention flag

When RAG_PERSIST_ANSWER_TEXT is left at its default false but RAG_ANSWER_CACHE_TTL_MS is enabled (default 300000), answered requests still call setSharedCachedAnswer and insert { answer: cloneAnswer(answer) } into rag_response_cache.payload (src/lib/rag-cache.ts:523-524). I only found lookup-time expires_at filtering/invalidation, not a purge that removes expired cache rows, so generated PHI-derived answer text can still remain at rest while this new flag and production-readiness check report answer persistence as off. A focused proof is to run any non-skipped answer request with the default cache TTL and inspect the inserted rag_response_cache payload.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[BigSimmo]

@codex

fix this issue below please

P1 Badge Gate cached answers behind the retention flag
When RAG_PERSIST_ANSWER_TEXT is left at its default false but RAG_ANSWER_CACHE_TTL_MS is enabled (default 300000), answered requests still call setSharedCachedAnswer and insert { answer: cloneAnswer(answer) } into rag_response_cache.payload (src/lib/rag-cache.ts:523-524). I only found lookup-time expires_at filtering/invalidation, not a purge that removes expired cache rows, so generated PHI-derived answer text can still remain at rest while this new flag and production-readiness check report answer persistence as off. A focused proof is to run any non-skipped answer request with the default cache TTL and inspect the inserted rag_response_cache payload.

[chatgpt-codex-connector]

Summary

• Gated shared persisted answer-cache reads and writes behind RAG_PERSIST_ANSWER_TEXT, preventing default TTL-enabled answer caching from writing generated answer text into rag_response_cache.payload at rest when retention is disabled. src/lib/rag-cache.tsL428-L435 src/lib/rag-cache.tsL516-L524
• Added focused Vitest coverage proving shared answer-cache payloads are not inserted when answer-text retention is disabled, and are inserted only after explicit opt-in. tests/rag-cache-invalidation.test.tsL68-L97 tests/rag-cache-invalidation.test.tsL129-L179
• Updated the privacy impact assessment so generated-answer storage and rag_response_cache retention documentation now match the new shared-cache retention gate. docs/privacy-impact-assessment.mdL60-L63 docs/privacy-impact-assessment.mdL236-L241 docs/privacy-impact-assessment.mdL383-L390
• Committed the changes on the current branch with commit 60d4708 fix(rag): gate shared answer cache persistence.
• Created the follow-up PR via make_pr with title: Fix shared answer cache retention gate.

Testing

• ⚠️ npm test -- --run tests/rag-cache-invalidation.test.ts initially could not run because node_modules was missing and the shell resolved Node 20 before setup.
• ✅ npm ci
• ✅ npm test -- --run tests/rag-cache-invalidation.test.ts
• ✅ npm run typecheck
• ⚠️ npm run check:production-readiness not run because repo instructions mark this provider/live-environment-adjacent check as confirmation-required.

View task →