Add three ChatGPT-style Answer result mockups

[BigSimmo]

Summary

The live Answer page is still a verified document card: AnswerResultSurface stages an AnswerCard with query echo, verification notice, support strength, prose, a sources capsule, and evidence sheets. Chat pieces already exist (UserQuestionBubble, prior-turn collapse), but the current result keeps card chrome, so the centrepiece does not read as a conversation.

This PR adds a design-scratch study at /mockups/answer-chat-redesign with three directions, all using the same clozapine-monitoring question and three sources. Shared mockup chrome is suppressed so each frame’s own composer is the only search bar.

• Simple — A message, not a card. Right-aligned question, left-aligned prose, numbered pills in the sentence, quiet source chips, in-thread quote preview.
• Sophisticated — The sentence lights the source. Superscript citations highlight the supporting claim. Desktop keeps a sources rail; phone uses an in-frame sheet. Hover/tap work both ways.
• Perfected — The source comes to the paragraph. A follow-up turn is already on screen. Footnotes expand an inline source card (desktop) or a compact in-frame sheet (phone). A Sources strip mirrors the marks. Copy/retry stay ghosted under the assistant turn.

Retrieval, ranking, and the production Answer surface are unchanged.

Verification

• Focused Vitest: tests/answer-chat-redesign-mockups.test.ts and tests/answer-chat-redesign-mockups.dom.test.tsx — 7 passed
• npm run verify:pr-local

Verification not run: verify:pr-local — mockup-only route, 404 in production; smallest gate that can fail is the focused mockup contract and citation-interaction tests above.

UI verification not run: verify:ui — no production UI, routing, or browser journey changed. Walkthrough screenshots/video of the mockup route follow in this session.

• 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
• npm run eval:retrieval:quality (must stay 36/36) when retrieval, ranking, selection, chunking, or scoring behavior changed
• npm run eval:rag -- --limit 15 + npm run eval:quality -- --rag-only when answer generation, the synthesis prompt, or answer post-processing changed
• 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

Risk and rollout

• Risk: Low. Design-scratch mockups under /mockups/** (404 in production unless mockups are explicitly enabled). No production Answer, retrieval, or citation-rendering code changed.
• Rollback: Revert the PR. No data or schema migration.
• Provider or production effects: None
• RAG impact: no retrieval behaviour change — mockups only; no files under src/lib/rag/, retrieval RPCs, golden fixtures, or ranking tests

Clinical Governance Preflight

• 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

These frames are synthetic design scratch. They do not change production answer generation, source rendering, or document access. Currency chips (current / review due) stay clinical, not decorative. The privacy line is the production obligation: “Do not enter patient-identifiable information.” plus the Privacy and data processing link.

Notes

Open /mockups/answer-chat-redesign (mockups enabled) to compare the three directions at phone 390 and desktop 1280. Citations are interactive.

[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 ↗︎.

[coderabbitai]

Important

Draft PR not reviewed

Draft PRs are not automatically reviewed by default.

• Trigger a manual review

To automatically review draft PRs, update your CodeRabbit configuration:

reviews:
auto_review:
drafts: true

---

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