Retire Therapy Compass parallel CSS into Clinical Sky tokens (PR-T)

docs/branch-review-ledger.md

@@ -575,10 +575,12 @@ Records before 2026-07-28 were written by hand and had drifted: 146 lines carrie
 | 2026-07-31 | claude/top-search-design-mockups-w53znc | c09eb64918ae2956968d35f3253c405ed9b915d5 | therapy-compass filter sheet + documents review fixes (C2) | Therapy Compass's phone Topics/availability selects were faking multi-select (value pinned to "", literal check-prefix in option text, a Clear-filters action among the options); replaced with one sheet of aria-pressed toggles matching the wide viewport, in the tc- idiom. Also lands two CodeRabbit findings from merged PR #1542: sheet footer primary action to sm:min-h-12, and the dialog test now pins aria-expanded/aria-controls. Third finding (unmount skips focus restore) dispositioned no-change with a test pinning the scroll lock is not leaked. | verify:cheap exit 0 (458 files / 4785 tests); ui-accessibility.spec.ts 15 passed chromium incl. the rewritten Therapy Compass case; ui-smoke documents journey 1 passed chromium |
 | 2026-08-01 | claude/top-search-design-mockups-w53znc | 4f4440fd6776742f5de203ee15295f372205321d | search results bar: scope-system deletion, filter shelf, bar anatomy | Handoff for PR #1555. Deleted the inert command-scope system (voided props, six modes' scope config, three matchers, four no-op call sites, the original shelf) — behaviour-preserving because every matcher early-returned true on a permanently-empty array. Rebuilt the applied-filter shelf prop-driven on live facet data, scoped to documents and therapy-compass. Landed the bar anatomy: tile spinner and funnel states, Filter to the right edge, Sort inboard. Study step 6 (remove the library button) deliberately declined — the nav route clears the query via onSearchModeChange. Ledger #182 closed. | verify:pr-local exit 0 (460 files / 4796 tests, production build, client-bundle secret scan, RAG fixtures 36 cases / 23 suites); ui-tools 87 passed; ui-smoke + ui-accessibility 108 passed 1 failed (pre-existing PDF-canvas test, fails identically stashed, Chromium 1194 vs pinned 1228); mutation-tested the shelf's survives-loading guard |
 | 2026-08-01 | claude/sentry-agent-monitoring-eri94v | 86983f344b45e42310e9f167a5adb0a56e46ddb5 | pr-1551 | merge-ready-pending-ci: merged origin/main; fixed outstanding-issues blank-line/#183 orphan + renumbered npm row to #204; kept worker+wizard error-tracking sections; qodo claim-spam thread already fixed on prior tip and resolved | check:outstanding-issues pass; merge-tree clean vs origin/main; prior tip Static PR failed on outstanding-issues; push 86983f344 |
-| 2026-08-01 | claude/ds-v2-tooling-loop | 40181192519fddc9405523d06bd3e691096cda6c | PR-0 tooling loop: Context7 + Chrome DevTools MCP wiring, design-sync and mockup-capture scripts, docs/env-example updates (11 files, +1005/-11, no clinical or RAG surfaces) | gates green | verify:pr-local: format/lint/typecheck/lock-parity green; unit 4803 passed, 6 env-class WSL relay failures in ci-cache-safety.test.ts (Ubuntu distro stopped), focused rerun 13/13 green after WSL boot; check:rag:fixtures 36 golden cases green; build skipped by selector |
+| 2026-08-01 | work | d4a4463172671cff3fd333feb685fa397e4f95d0 | latency, lazy loading, React and Next.js frontend performance | P2 eager document administration bundle boundary fixed; no P0/P1 findings | lint; typecheck; 463 files/4811 tests; production build; bundle budget |
 | 2026-08-01 | codex/cloud-connected-profile-boundary | 6fddcfc780b237b6b2cd524dbebd7b1de70d9701 | Cloud connected profile credential boundary | No high-confidence issues after least-privilege MCP hardening and portable Git fixture fix | Cloud static PASS; focused Vitest 15/15; full format PASS; Bash syntax PASS; PR-local dry-run |
 | 2026-08-01 | codex/cloud-connected-profile-boundary | 55b08496a5ee3495eed8a7436e8f69ae7b6612d8 | Cloud connected profile credential boundary | Reviewer findings fixed: cross-tenant service-role credential scrubbed and duplicate Supabase MCP parameters rejected | Cloud static PASS; focused Vitest 15/15; Bash syntax PASS; targeted Prettier PASS |
-| 2026-08-01 | work | d4a4463172671cff3fd333feb685fa397e4f95d0 | latency, lazy loading, React and Next.js frontend performance | P2 eager document administration bundle boundary fixed; no P0/P1 findings | lint; typecheck; 463 files/4811 tests; production build; bundle budget |
 | 2026-08-01 | codex/review-latency-and-lazy-loading-optimizations | 5f069a7fec4e6ada47a0074aa5f2ea2c9dc97830 | pr-1562 unblock | unblocked: merge-tree was clean behind-by-4; merged origin/main; no unresolved threads; prior tip CI green including Production UI + PR required | merge-tree clean vs origin/main; gh mergeable was CONFLICTING/DIRTY (staleness); unresolved threads 0; auto-merge off |
+| 2026-08-01 | claude/ds-v2-therapy-teardown | f9cf75fefcf78a2049fa756322265ecb64845254 | PR-T therapy-compass CSS teardown: delete therapy-compass.css, migrate tc-* to design-system control recipes, resolve #205/#016(e); focusRing local after ui-primitives rename | gates green | check:design-system-contract green; docs-surface (links/scripts/inventory/index) green; test:e2e:critical 15/15; verify:ui 344 passed; verify:pr-local green (build+client-bundle+rag fixtures 36/36) |
+| 2026-08-01 | claude/ds-v2-therapy-teardown | 4c96b90aa3e99f3e3aec81a005a1da6bf7a8792f | PR-T therapy-compass CSS teardown: delete therapy-compass.css, migrate tc-* to design-system control recipes, resolve #205/#016(e); focusRing local after ui-primitives rename | gates green | check:design-system-contract green; docs-surface (links/scripts/inventory/index) green; test:e2e:critical 15/15; verify:ui 344 passed; verify:pr-local green (build+client-bundle+rag fixtures 36/36); tip includes ledger append |
+| 2026-08-01 | claude/ds-v2-tooling-loop | 40181192519fddc9405523d06bd3e691096cda6c | PR-0 tooling loop: Context7 + Chrome DevTools MCP wiring, design-sync and mockup-capture scripts, docs/env-example updates (11 files, +1005/-11, no clinical or RAG surfaces) | gates green | verify:pr-local: format/lint/typecheck/lock-parity green; unit 4803 passed, 6 env-class WSL relay failures in ci-cache-safety.test.ts (Ubuntu distro stopped), focused rerun 13/13 green after WSL boot; check:rag:fixtures 36 golden cases green; build skipped by selector |
 | 2026-08-01 | claude/ds-v2-tooling-loop (PR #1568) | f6317ea48355c26e080ce45b6f574159e8397e24 | PR #1568 unblock/fix | merged origin/main (behind-but-clean; merge-tree was clean); cleared GitHub DIRTY/CONFLICTING; advisory review threads left open (not merge/CI blockers); auto-merge remains armed | merge-tree clean; ahead 7 behind 0; no provider-backed checks; no product code fix needed |
 | 2026-08-01 | claude/ds-v2-tooling-loop (PR #1568) | b76bcc9fbc39f2f986a52869c810ba3cb89994f2 | PR #1568 review-thread resolve | fixed and resolved all 13 review threads (fail-closed project identity, inventory exit code, provenance demo id, buildCmd execution, chrome pin, context7 rollback); ledger row for superseded 93d41c1 dispositioned | node --check scripts; design-sync --dry-run; no provider-backed checks |

docs/outstanding-issues.md

@@ -130,7 +130,7 @@ removed after current-main verification; it is not missing recommended work.
 | 81 | `#204` | A2 | High — install/CI integrity | Next dependency or verify:pr-local change | 1–3 hours | npm 11.6.2 regenerates a lockfile its own `npm ci` rejects, reddening every CI job — add `npm ci --dry-run` (or lockfile-sync assertion) to `verify:pr-local` when package.json/lock change; do not regenerate with `npm install` to "fix". Distinct from #149. |
 
 
-<!-- issues:next-id=205 -->
+<!-- issues:next-id=206 -->
 ## Open items
 
 > **Merged-main canary update (2026-07-23, run `30018289898`):** the new structured report correctly recorded evaluated tree `c24f2e8f2d30d0c59fc1eba025d3dcd63478137e`, run/attempt identity and `cross-region-runner` latency context. Golden retrieval remained 36/36 with document/content recall 1.0 and no failed cases. The 44-case answer gate had grounded-supported and unsupported-correct rates of 1.0, but failed because `neuroleptic-side-effect-escalation` again returned one citation where two are required (citation-failure rate 0.0227). `admission-discharge-comparison` again omitted the specific AKG admission document after `comparison_source_extractive_fallback`; `admission-discharge-coverage-paraphrase` was advisory-only at 24,870 ms. Answer cost was reported as `$0.234736`. Do not retry immediately: retain this as the first structured datapoint, compare it with the scheduled 2026-07-26 report, and keep retrieval/ranking unchanged.
@@ -154,7 +154,7 @@ removed after current-main verification; it is not missing recommended work.
 | #057 | P2 | task | Complete staging soak and rollback rehearsal | After #056, run the documented soak and rollback against an exact candidate; retain latency/error/rollback evidence. Stop on unsafe data, identity mismatch, or an unowned rollback decision. | `docs/launch-operator-runbook.md`; `docs/capacity-review.md` | 2026-07-24 |
 | #011 | P3 | task | Auth DB-connection allocation is operator-only | Supabase Auth (GoTrue) is capped at ~10 absolute DB connections (Supabase perf advisor). Switch to **percentage-based** allocation in the Supabase **dashboard** before the first compute scale-up — **not settable via SQL/MCP** (operator-owned). Verify via a staging soak + an approval-gated read-only advisor re-check. | `docs/auth-connection-cap-runbook.md`; `docs/process-hardening.md` (Known follow-up debts) | 2026-07-21 |
 | #013 | P3 | rec | Route-chunk + mockup catalogue JSON weight | Keep this recommendation open and measurement-gated. `build:analyze` still finds route-scoped catalogue modules: `/specifiers` ships `specifiers-search-index.json` (~180 KB parsed), `/forms` ships `forms-catalog.json` (~132 KB), and `/formulation` ships `formulation-content.json` (~52 KB; client-side local search needs an index/full split or a search endpoint). The approved Lighthouse sample covered `/forms`, but `/specifiers` and `/formulation` remain unmeasured, so the precommitted `#017` rule does not permit archiving their payload work. Development-only `*-mockups.tsx` chunks are not an initial production bundle and production returns 404 for `/mockups/*`; do not restructure them without deploy-artifact or cold-start evidence. **Next:** collect route-specific LCP/CLS and CrUX INP evidence for `/specifiers` and `/formulation`, then close only the routes that meet every `#017` threshold. | session 2026-07-21 (`build:analyze`); PR #1470 review | 2026-07-21 |
-| #016 | P3 | rec | "Big but not easy" structural + motion perf | Deferred larger levers: (a) nonce-CSP forces every product route to `╞Æ Dynamic` (zero static generation) — evaluate Partial Prerendering / static shells for the static clinical catalogues (DSM/differentials/therapy/specifiers/formulation); (b) sidebar expand/collapse animates `grid-template-columns` (biggest smoothness cost, motion-gated — needs a transform-overlay rethink); (c) Therapy Compass fetches 692 KB / 2.5 MB JSON client-side (defer until interaction + confirm brotli); (d) settings/setup/admin dialogs static-imported into the home chunk (`next/dynamic` them); (e) `src/app/(search-app)/layout.tsx:4` imports 71.6 KB of Therapy-Compass-only CSS in the ROUTE-GROUP layout, making it render-blocking on `/`, `/documents`, `/forms`, `/dsm` and every mode home; (f) `shared-search-app-shell.tsx:8` statically imports the `therapy-compass` barrel, pulling `workspace.tsx` + `bindings.tsx` + `nav.tsx` into every `(search-app)` route; (g) three client waterfalls (`use-app-preferences.ts:156-182`, `ClinicalDashboard.tsx:977-1069`, `signed-image.tsx:60-84` + `use-signed-image-url.ts:39`) and the paint offenders in `globals.css` beyond the sidebar grid — three stacked `backdrop-filter` passes on an always-mounted translating element (`:709-748`), `box-shadow` inside a `transition` list (`:677-684`), and `@keyframes shimmer` animating `background-position` on the shared `Skeleton` (`:2289-2296`). **CORRECTED 2026-07-29 on (c):** the Therapy Compass filenames are unversioned and Next serves `/public` with an ETag, so only the FIRST visit pays 690.6 KB / 2,470 KB — repeat visits pay ~4 revalidation round trips. The fix is content-hashed filenames + `immutable` (touching `scripts/build-therapies-index.mjs` and `check:therapy-data-index`), NOT a bare `Cache-Control` line. See `docs/audit/latency-audit-2026-07-28.md` L3-1/L3-2/L3-3/L3-6/L3-7. | session 2026-07-21 (build route table + design audit) | 2026-07-21 |
+| #016 | P3 | rec | "Big but not easy" structural + motion perf | Deferred larger levers: (a) nonce-CSP forces every product route to `╞Æ Dynamic` (zero static generation) — evaluate Partial Prerendering / static shells for the static clinical catalogues (DSM/differentials/therapy/specifiers/formulation); (b) sidebar expand/collapse animates `grid-template-columns` (biggest smoothness cost, motion-gated — needs a transform-overlay rethink); (c) Therapy Compass fetches 692 KB / 2.5 MB JSON client-side (defer until interaction + confirm brotli); (d) settings/setup/admin dialogs static-imported into the home chunk (`next/dynamic` them); (e) **DONE 2026-08-01 in PR-T (ds-v2 therapy teardown):** deleted `therapy-compass.css` and removed its route-group layout import — no longer render-blocking on `/`, `/documents`, `/forms`, `/dsm` and every mode home; (f) `shared-search-app-shell.tsx:8` statically imports the `therapy-compass` barrel, pulling `workspace.tsx` + `bindings.tsx` + `nav.tsx` into every `(search-app)` route; (g) three client waterfalls (`use-app-preferences.ts:156-182`, `ClinicalDashboard.tsx:977-1069`, `signed-image.tsx:60-84` + `use-signed-image-url.ts:39`) and the paint offenders in `globals.css` beyond the sidebar grid — three stacked `backdrop-filter` passes on an always-mounted translating element (`:709-748`), `box-shadow` inside a `transition` list (`:677-684`), and `@keyframes shimmer` animating `background-position` on the shared `Skeleton` (`:2289-2296`). **CORRECTED 2026-07-29 on (c):** the Therapy Compass filenames are unversioned and Next serves `/public` with an ETag, so only the FIRST visit pays 690.6 KB / 2,470 KB — repeat visits pay ~4 revalidation round trips. The fix is content-hashed filenames + `immutable` (touching `scripts/build-therapies-index.mjs` and `check:therapy-data-index`), NOT a bare `Cache-Control` line. See `docs/audit/latency-audit-2026-07-28.md` L3-1/L3-2/L3-3/L3-6/L3-7. | session 2026-07-21 (build route table + design audit) | 2026-07-21 |
 | #018 | P2 | task | Split the lithium, ADHD and metabolic residuals by mechanism | Current evidence keeps the mechanisms separate. **Lithium — closed within this item:** the row/atom-aware subject guard, foreign-parameter rejection and query-specific range promotion returned `0.5–1.0 mmol/L` with correct targeting/citation; the full retrieval canary remained 36/36 with recall 1.0 and zero per-case RR regressions, and the full answer canary passed every blocking gate. **ADHD — open corpus debt:** `CG.MHSP.ADHD.pdf` is absent from the hosted corpus and the retrieved chart exposes `accessible_table_count=0`; repair corpus/fixture or ingestion evidence rather than weakening extractive budgets. **Metabolic — open structured-evidence debt:** the standalone plural classifier worsened the live answer and was reverted; obtain auditable schedule text/table evidence before another candidate. | targeted live lithium/ADHD/metabolic evidence 2026-07-27; `docs/evidence/rag-reliability-evidence-2026-07-27.md`; refuted approaches | 2026-07-21 |
 | #022 | P2 | task | Source-governance metadata refresh (operator) | The selected policy is now encoded locally as auditable `third_party_reference_attested` evidence with policy version, reviewer qualification, evidence references and append-only review history. It deliberately preserves `clinical_validation_status=unverified`; malformed, stale or non-BMJ evidence remains review debt. Migration `20260727010000_bmj_third_party_source_attestation.sql` is prepared but was **not applied**. The ten most visible local-document candidates are captured in `docs/evidence/rag-top-local-review-manifest-2026-07-26.json` with `attestation_applied=false`; qualified human review, deliberate hosted apply/attestation, and warning-rate remeasurement remain operator work. | governance worklist; local policy/migration tests; top-ten evidence manifest | 2026-07-21 |
 | #023 | P2 | task | Complete scheduled browser and labeling disposition | **Partial 2026-07-30:** `release-browser-matrix` no longer depends on `pr-required`, so a blocking scheduled dependency audit cannot skip Firefox/WebKit. Still need one green matrix datapoint + human irrelevant-at-10 disposition. The 2026-07-26 retrieval and answer artifacts are read and compared under resolved #051. Scheduled CI run `30216361999` failed its existing production dependency audit before Firefox/WebKit, while production Chromium passed. After that audit is green, capture one scheduled/manual browser-matrix datapoint; separately record the human decision for the stable irrelevant-at-10 set. #084 now makes each top-10 grade and matched signal reproducible, but it does not substitute for the human disposition. Do not rerun or spend on RAG for this item. | runs `30216191889`/`30216361999`; per-rank diagnostics #084; session 2026-07-27 | 2026-07-21 |
@@ -362,5 +362,6 @@ Move resolved rows here with the resolution date and a one-line outcome. Keep th
 | #185 | task | Confirm hosted migration history row for 20260731150000 db-query-perf | Confirmed on live Clinical KB Database (sjrfecxgysukkwxsowpy): supabase_migrations.schema_migrations has version 20260731150000 name db_query_perf_rate_limit_and_image_indexes; document_images_searchable_doc_page_relevance_idx present. No re-apply needed. | 2026-07-31 |
 | #184 | task | Regenerate supabase/drift-manifest.json after db-query-perf migration | Regenerated supabase/drift-manifest.json via Docker npm run drift:manifest on host; includes document_images_searchable_doc_page_relevance_idx and updated consume_api_*_rate_limit inventory; schema_sha256 matches schema.sql. | 2026-07-31 |
 | #182 | issue | The command-scope system is inert — chips are configured and matchers exist, but nothing can ever set a scope | Deleted. Took the option the row recommended: dropped the two voided props from universal-search-command-surface, the scope state and callbacks from both shells and the master header, the scopes config arrays and CommandScopeChip type, the three matcher functions, the four no-op consumer filters (services, forms, prescribing, favourites), and the band's scope shelf plus its empty-state Clear-scopes button. Net 453 deletions / 23 insertions across 13 files. Behaviour-preserving by construction: every matcher early-returned true on the permanently-empty array, so each call site was already a no-op. SearchCommandContext survives with query/modeId — applications-launcher-page still consumes it. Verified verify:cheap exit 0 (460 files / 4792 tests) and ui-smoke 93 passed, the one failure being the pre-existing PDF-canvas test that fails identically without this change. | 2026-07-31 |
+| #205 | task | Retire therapy-compass.css (tc- class system) in favour of design-system classes | Resolved in PR-T (claude/ds-v2-therapy-teardown): therapy-compass.css deleted, all tc- classes migrated to design-system classes (therapyBtn/outlineControl/softControl), layout import removed, filter-sheet migrated; design-system contract check green. qtc- (cardiac QTc) identifiers deliberately untouched. | 2026-08-01 |