SoundMindsAI · SoundMindsAI · May 17, 2026 · May 17, 2026
@@ -2,13 +2,13 @@
 
 > Read this first. Snapshots the active branch, what just shipped, what's in flight, what's queued, and where the project currently sits in the MVP1 → GA roadmap. Updated whenever a feature lands or a priority shifts.
 
-**Last updated:** 2026-05-17 (after `infra_e2e_seed_completed_study` merged as PR #130 — dev-only `POST /api/v1/_test/studies/seed-completed` endpoint + `seedStudyCompletedWithDigest` Playwright helper enabling deterministic E2E coverage of completed-study surfaces. Strict env guard returns 404 outside `ENVIRONMENT=development`. 2 commits-of-followup captured `infra_e2e_wire_seed_helper_into_studies_spec/idea.md` for the deferred Playwright tests that broke the smoke lane on first push).
+**Last updated:** 2026-05-17 (after `chore_data_table_primitive_followups` items 1+2+4+6 of 6 merged as PR #132 — test mock helper extraction + `useLocalStorageSet` docstring lock + `?limit=` allowlist coercion + URL-state validation in `useDataTableUrlState`. Items 3 + 5 remain open per the chore's idea file. UI vitest **368 passing across 56 files** — +11 cases. Defense-in-depth: direct URLs with `?status=invented` or `?sort=garbage:asc` now hydrate as empty matchers instead of being sent to the backend and 422-ing).
 
 ---
 
 ## Current branch / execution context
 
-- **Branch:** `main` (post-merge of PR #130 squash commit `13b3383` 2026-05-17). Finalization docs PR off `docs/finalize-e2e-seed-completed-study`. Prior squashes: PR #128 `73459d2` 2026-05-17 (bug_cursor_decode_value_validation), PR #126 `d6115b3` 2026-05-16 (feat_data_table_primitive). `v0.1.0` annotated tag still on `main` commit `d099536` 2026-05-13; GitHub Release at https://github.com/SoundMindsAI/relyloop/releases/tag/v0.1.0.
+- **Branch:** `main` (post-merge of PR #132 squash commit `ee4c8d4` 2026-05-17). Finalization docs PR off `docs/finalize-data-table-followups`. Prior squashes: PR #130 `13b3383` 2026-05-17 (infra_e2e_seed_completed_study), PR #128 `73459d2` 2026-05-17 (bug_cursor_decode_value_validation), PR #126 `d6115b3` 2026-05-16 (feat_data_table_primitive). `v0.1.0` annotated tag still on `main` commit `d099536` 2026-05-13; GitHub Release at https://github.com/SoundMindsAI/relyloop/releases/tag/v0.1.0.
 - **Active feature:** none in flight. **Three PRs shipped 2026-05-15:** PR #122 (Phase 1, 16th MVP1 feature — Tooltip primitive + 26 placements on create-study modal + study detail), PR #123 (Phase 1 finalization docs), PR #124 (Phases 2 + 3 — 17th MVP1 feature; 21 additional tooltips on judgments + proposals + cluster registration + 2 new first-run components: chat ExamplePrompts strip + Stripe-style StartHereChecklist on home page). The original "MVP1 Phase 1 only" scope-lock was reversed mid-day: operator decided to ship Phases 2 + 3 together with a Stripe-style design call rather than wait for MVP2. PR #124 took 2 hours from idea-folder reuse to merge. 47 total tooltip placements + 2 new first-run components live in `main`. **PR #122 shipped 2026-05-15 morning** — `feat_contextual_help` Phase 1 (16th MVP1 feature). Adds the first Tooltip primitive (`@radix-ui/react-tooltip@~1.2.8` + shadcn-style wrapper at `ui/src/components/ui/tooltip.tsx`), two glossary-backed wrappers (`InfoTooltip` standalone + asChild modes; `HelpPopover` click-to-open with `react-markdown` safety filter), and a 49-key glossary source-of-truth at `ui/src/lib/glossary.ts` (8 enum groups parity-tested against `enums.ts`). 26 tooltip placements across the create-study modal (Step 1 target + Step 3 template + 9 Step 5 inputs), study-header (status badge dynamic key + Best metric + Trials), trials-table (5 column headers + Sort label), and digest panel (5 section labels + Open PR enabled + Open PR disabled). The disabled Open PR button refactored from native `disabled` to `aria-disabled="true"` so it stays focusable and the tooltip reveals on focus (AC-11). Gemini Code Assist: 2 findings (1 accepted + fixed, 1 rejected with cited counter-evidence). Final GPT-5.5 review: 1 Medium accepted-framing-but-deferred. Spec converged at GPT-5.5 cycle 3 (24 findings, 23 accepted + 1 rejected); plan converged at cycle 2 (12 findings, 10 accepted + 1 rejected + 1 spec patch). UI vitest now **279 passing across 48 files** (was 249 across 45 — +3 new test files, +30 cases). Playwright E2E **8 passing** (was 5 — +3 new contextual-help tests). One follow-up filed: `infra_e2e_seed_completed_study/idea.md` tracks the E2E gap for digest-panel triggers + AC-11 (cross-subsystem helper for seeding a completed study with digest + proposal; component-level coverage is in place). Phases 2 + 3 deferred to MVP2 via `feat_contextual_help_mvp2/` (judgments + proposals tooltips; chat + cluster + home onboarding; the home-page "Start here" panel is the only product-design-shaped item).
 
 **Earlier — seven PRs shipped 2026-05-14:** `feat_judgments_periodic_resume_sweep` (PR #104, 14th MVP1 feature), `bug_query_inline_crud_since_filter_uuidv7_ms_collision` (PR #106 — UUIDv7 ms-collision test flake), `infra_dashboard_regen_pre_commit_conflict §2+§4` (PR #108 — dashboard regen idempotency + relative-link rewriting), `infra_make_targets_split_backend_only` (PR #110 — `make backend-fmt/lint/typecheck` + symmetric `ui-fmt` so Node-18 contributors aren't blocked), `chore_digest_worker_narrow_except` (PR #112 — narrowed `except Exception` allowlist to `(ValueError,)` + ERROR-level `digest_importance_failed_unexpected` event), `infra_structlog_test_helpers` (PR #114 — factored the two structlog test-assertion patterns into `backend/tests/_log_helpers.py`), and `chore_chat_last_message_preview` (PR #117 — `last_message_preview` + `last_message_at` on `ConversationSummary` via LATERAL JOIN; frontend shows preview under title + swaps displayed timestamp from `created_at` to `last_message_at`). Plus PR #116 dropped `chore_studies_ui_shadcn_polish` as won't-do (forward-compat audit on NavigationMenu primitive + ClusterFilterSelect precedent on native `<select>` for page-level controls). The 7-PR day drained the operational-friction surfaces and the last `/bug-fix`-shaped item. **MVP1 alpha + 17 features shipped (all 3 phases of feat_contextual_help on 2026-05-15) + 26 backlog items drained.** Manual maintainer steps still pending from [`release-checklist.md`](docs/03_runbooks/release-checklist.md): §4 fresh-VM hosted-OpenAI walkthrough, §5 local-LLM walkthrough, §8 feedback Discussion + design-partner channel shares. **Remaining MVP1 backlog: zero actionable items** — the MVP1 dashboard now reads `Path to MVP1: 0`. Two ideas held for MVP2 (visible on the new `MVP2_DASHBOARD.md`): `infra_arq_subprocess_test_mvp2` (folder renamed 2026-05-14 to mirror the `_mvp2` precedent — trigger-locked: arq pin bump, 3rd cron, or MVP3 hardening opt-in) and `bug_chat_long_conversation_truncation_mvp2` (paired with the just-shipped chat preview but kept on MVP2 hold per its preflight — needs `/pipeline` for migration + new LLM round-trip). Plus 4 keep-deferred items by operator decision + `infra_dashboard_regen_pre_commit_conflict §3` follow-up (runbook addendum, MVP1-eligible but low-leverage).
@@ -19,6 +19,8 @@
 
 ## Most recent meaningful changes (newest first)
 
+- **2026-05-17 — `chore_data_table_primitive_followups` items 1+2+4+6 merged into `main`** as PR #132 (squash commit `ee4c8d4`). Shipped via direct implementation → `/impl-execute --ad-hoc`. 4 of 6 follow-ups from `feat_data_table_primitive` (PR #126) review cycles. **Item 1 (test helper extraction):** new `ui/src/__tests__/helpers/data-table-url-mock.ts` factors the ~25-line `searchParamsSubscribers` + `applyUrl` mock pattern duplicated across 3 test files (`proposals/page.test.tsx`, `judgments/[id]/page.test.tsx`, `query-sets/queries-table.test.tsx`). Uses async `vi.mock` factory + dynamic import to sidestep Vitest's hoist trap (a vi.mock factory that references same-file imports fails because hoisting puts the factory above the import statements). **Item 2 (`useLocalStorageSet` shape lock):** plan §Story 2.10 had sketched `{ value: string[], ... }`; shipped impl uses `Set<string>` for O(1) membership + `.has()` / `.toggle()` matching the consumer pattern. Docstring updated to canonicalize the shipped contract; no code change. **Item 4 (`?limit=` allowlist):** when `options.pageSizeOptions` is provided, out-of-allowlist `?limit=` values fall back to `defaultPageSize`. Backward-compatible — current consumers omit `pageSizeOptions`. **Item 6 (URL-state validation):** centralizes validation for enum filters (`column.filter.wireValues`) and sort tokens (`column.sortable` + `column.sortDirections`) inside `useDataTableUrlState`. Direct URLs like `?status=invented` or `?sort=garbage:asc` now hydrate as empty matchers instead of round-tripping to the backend and 422-ing. fk-select filters pass through (their option IDs load async — can't be hook-validated). Did NOT auto-clean URLs via `replace()` — silently drop to avoid the "flash of empty state" UX surprise. **Gemini Code Assist:** 2 Medium findings, both accepted: (1) docstring on sort fallback said "first allowed direction" but impl hardcoded `'asc'` — rewrote docstring to lock in the deliberate "no synthesizing direction" semantic (mirrors backend `parse_sort` at `_sort.py:50-56`) + regression test for desc-only edge case; (2) `Number(limitRaw)` → `parseInt(limitRaw, 10)` so fractional URLs like `?limit=10.5` truncate to `10` + regression test. **CI:** all 7 checks green on both cycles (first commit + Gemini fixes). **Tests:** UI vitest 357 → 368 (+11 cases — 7 for item 6 validation, 4 for item 4 coercion). Lint + typecheck clean. **Items 3 (`DataTableProps` URL state aggregate prop) + 5 (TanStack `state.columnVisibility` wire-up) still open** in the chore's idea.md — both touch 8 DataTable consumers each; too big for one ad-hoc PR.
+
 - **2026-05-17 — `infra_e2e_seed_completed_study` merged into `main`** as PR #130 (squash commit `13b3383`). Shipped via direct implementation → `/impl-execute --ad-hoc` (no `/bug-fix` because this is `infra_`-prefixed, not a bug). Adds a development-only endpoint that inserts a completed study + 2 trials + digest + (optional) pending proposal in one transaction so future Playwright E2E coverage of digest-panel surfaces can run against real backend rows instead of mocked component data. **Endpoint:** `POST /api/v1/_test/studies/seed-completed`. **Security model:** new `Settings.environment: str = "development"` field; endpoint gates on `environment == "development"` (strict equality — not `!= "production"`) and returns 404 `RESOURCE_NOT_FOUND` outside development so the staging/production behavior is indistinguishable from "route not registered." Pydantic `extra="forbid"` on the request schema blocks tampered payloads from smuggling extra columns into the insert path. **Implementation:** new `backend/app/api/v1/_test.py` router with `_require_development_env` dependency, new `backend/app/services/test_seeding.py` service helper that drives the study through legal state-machine transitions via `study_state.start_study → complete_study` so the FR-7 protection listener does not raise, plus `seedStudyCompletedWithDigest` Playwright helper at `ui/tests/e2e/helpers/seed.ts`. **Tests:** 13 contract cases in `backend/tests/contract/test_test_endpoint_guard.py` covering env-guard HTTP-layer + dependency-layer behavior across 5 non-dev environment values + schema validation; 2 integration smoke cases in `backend/tests/integration/test_test_seeding.py` exercising the live-DB write path; `# pragma: no cover` on the integration-only function bodies (matching the precedent from `feat_github_pr_worker` PR #45 `git_pr.py`). **Gemini Code Assist:** 2 Medium findings, both accepted + fixed in commit `b617ca1` (`timedelta` import + reorder `start_study` BEFORE trial inserts so the seeded flow mirrors the real orchestrator + anchor trial timestamps off `study.started_at` with `ended_at = started_at + timedelta(milliseconds=duration_ms)`). **CI fight:** 5 push cycles before green — (1) initial push `580edf1`: backend coverage gate failed at 80% because the new ~290 LOC weren't covered; (2) `b617ca1` Gemini fixes: same coverage gate; (3) `6964924` added integration test: still failing because... (4) `a68007b` added `# pragma: no cover` to integration-only functions: STILL failing — turned out the root cause was actually mypy `--strict` errors in `test_test_endpoint_guard.py` (two issues: `test_router.HTTPException` was importing through a re-export that didn't exist, and `assert isinstance(detail, dict)` was unreachable because FastAPI types `HTTPException.detail` as `str`); diagnostic blind spot was running mypy on touched files only instead of `mypy backend/` per CLAUDE.md "Build, Test, and Lint Commands"; (5) `2615aae` fixed the mypy errors — backend lane green, smoke lane then failed because the 2 new Playwright tests in `studies.spec.ts` broke `make up`-driven E2E; root cause undiagnosed (agent execution environment has no authenticated access to GitHub Actions logs or Playwright report artifacts); dropped both tests + captured resume work as `infra_e2e_wire_seed_helper_into_studies_spec/idea.md` in commit `3203787`. **Final state:** all 7 CI checks green; endpoint + helper + integration + contract tests ship as primary deliverable; the 2 dropped E2E tests are additive coverage that future session with log access can restore in <60min via the captured idea (the helper, endpoint, and seed function are all already wired correctly — just need to debug the Playwright failure pattern). **Architecture doc impact:** none; `Settings.environment` documented inline at `backend/app/core/settings.py:295-302` referencing CLAUDE.md §Environments.
 
 - **2026-05-17 — `bug_cursor_decode_value_validation` merged into `main`** as PR #128 (squash commit `73459d2`). Shipped via `/bug-fix` (Default mode) → `/impl-execute --ad-hoc`. Defense-in-depth fix on [`backend/app/db/repo/_sort.py`](backend/app/db/repo/_sort.py) `decode_cursor` — adjudicated the Medium GPT-5.5 finding deferred from PR #126 final review. **Root cause:** decode performed `json.loads → decoded[0] → str(decoded[1])` with no shape or type validation. A tampered cursor whose value-half was a non-primitive (dict / list / bool) silently passed through when `value_is_datetime=False`, flowed into `keyset_predicate`'s SQL comparison, and surfaced as 500 INTERNAL_ERROR instead of the intended 422 VALIDATION_ERROR that all 9 cursor-consuming routers wrap `_sort_decode_cursor` to produce. **Fix:** assert 2-element list payload; assert value-half is `None | str | int | float` (bool rejected via explicit isinstance check since `isinstance(True, int)` is True); assert row-id is str (Gemini High finding — `str(decoded[1])` would coerce `None` → `"None"` / `[]` → `"[]"` and surface as 500 on UUID id columns). All decode failures normalize to `ValueError`; routers' existing `except Exception` translation fires unchanged. **Locked-and-deferred design forks:** (1) skip per-column-type cross-check (4-type allowlist already covers every legitimate value); (2) keep 2-tuple cursor payload — defer 3-tuple `[col_name, value, row_id]` for cross-sort replay detection (backward-incompat with issued cursors, addresses non-crash concern); (3) reuse `VALIDATION_ERROR` code instead of adding `INVALID_CURSOR` (9-router precedent; new code would add noise without changing client behavior). **Tests:** +12 unit cases in `backend/tests/unit/db/test_parse_sort.py` (7 for the original bug surface — dict / list / bool value-half, single-element / three-element / non-list payload, int-with-datetime-sort — plus 5 parametrized for Gemini's row_id finding covering None / int / list / dict / bool). All 12 verified to fail on `main` (proven via `git stash`) and pass with the fix; broader 855 unit + 157 contract suite green. **Gemini Code Assist:** 3 findings, all 3 accepted + fixed (High row-id type check + Medium row_id regression test + Medium stale bug_fix.md line ranges). **GPT-5.5 final review:** skipped — agent session has no `.env`/`OPENAI_API_KEY` in remote execution environment; diff is small (+192/-4 across 3 files) and bounded to a single helper module so recommended-but-not-required per impl-execute Step 7 threshold. **CI:** both push cycles green on first attempt (no flakes); 7 checks each cycle. **Scope discipline:** no router edits, no migration, no wire-format change, no `api-conventions.md` change — pre-fix and post-fix cursors decode identically for legitimate payloads.