QA audit + safe bug fixes + $30 community tier spec - #32
Open
ethanwolk04-maker wants to merge 5 commits into
Open
QA audit + safe bug fixes + $30 community tier spec#32ethanwolk04-maker wants to merge 5 commits into
ethanwolk04-maker wants to merge 5 commits into
Conversation
…in-status comparison Adds docs/qa: full coach+client QA audit of the app (68 bugs), flows to redesign, a subsystem primer, and a re-validation of every bug against the shipped main (12 already fixed, 3 fixed-in-effect, 43 still open). Also fixes B23: the live workout player was hardcoding rest_sec=90 for every coach-assigned exercise, discarding the coach's prescribed rest (strength mains are prescribed 240-300s). coachingWorkoutToPlan now honors ex.rest_sec when present, falling back to 90s only when none was set. Engine intentionally untouched (goal->reps and 2-working-sets are product decisions per Ethan). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Applies 23 low-risk, reviewable fixes from the QA audit (engine untouched). Python services syntax-checked (py_compile); TS/RN hand-verified, pending CI. Backend (apps/api): - B18/B56/B58/B59 coaching_cohort: at-risk + unread aggregate + weekly check-ins no longer scoped to the first page; per-client unread is a real count not 1. - B38 corpus_db: paginate the embedded-id fetch (was capped ~1000, re-embedding). - B39 corpus_pipeline: batch embeddings <=2048 inputs (was one 3000-input call). - B41 corpus_chat: prune citations to the [N] markers the answer actually cites. - B33 coaching_template_adapter: drop the misleading weak-point note (2-set kept). - B34 coaching_program_validator: add day-name readability validation. Mobile (apps/mobile): - B05 coach card renders real set count (numeric-string fallback), B35 shows current/keep split, B36 AboutTab reads primary_goal. - B09 don't save a 0-set session; B31 roll back the optimistic record on failure; B55 fix coachingTabMe spread clobbering linked:true. - B22 drop the redundant "building your plan" modal (keep inline); B45 scope the dismissed flag to the revision. - B30 leaderboard uses real participant/user counts, not hardcoded 1,000/20,000. - B07 full-gym prefill no longer also sets bodyweight_only; B10 dedupe coach recent-workouts across Program/App; B61 A/B labels for repeated leg days. - B52 nutrition load sequence guard; B53 gate hub poll on app active; B28 reword Progress "check-in" copy to measurements. Docs: docs/qa/FITFO_TIER_30_COMMUNITY_SPEC.md — real build spec for the $30 community tier (not a paste of the client-side demo mock). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
… checkpoint) Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Restructure per expert panel: lead with promise, prove with the coaches (on-record moved up), show the rhythm, show the app, VIP-first two-card pricing, side-by-side, repeated closing CTA, FAQ. CTAs sell the next step (Start my plan) with price underneath. Header/hero decluttered, IG moved to footer. Zero em dashes. Features, prices, and the no-auto-renew promise kept exactly. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The /demo page was a preview: state lived in the browser tab, so nothing survived a refresh, a new tab or a second device. Nobody could be a member of it. This makes the tier real at /community, reusing what already exists rather than inventing anything. Real accounts. Sign up / sign in go through the existing account system (POST /auth/email/signup, /auth/email/login) that the mobile app uses, so a Community member is a normal FitFo account with the same 30 day JWT. Real saved progress. A new ct_member_state table (migration 196) stores the member's session blob, restored on sign in and mirrored back on every change (debounced 1.5s, flushed on unmount). The app itself is untouched. DemoApp still reads and writes its session exactly as before; the /community wrapper seeds that session from the server before mount and subscribes to writes, via two small additions to demoSession.ts. /demo has no subscribers, so its behaviour is unchanged and it remains a public, no login preview. Naming: the tier owns the `ct_` prefix, kept separate from coaching_* so the two tiers never collide. Verified: py_compile clean (community_tier.py, main.py); apps/web tsc --noEmit zero errors with all 7 new/changed files in the compiled file list; apps/web production build succeeds and registers /community, /api/community-tier/auth and /api/community-tier/state. Not verified: the end to end signup flow was not run against production, which needs live credentials and would create real accounts. The runbook at docs/qa/FITFO_TIER_30_BACKEND_RUNBOOK.md is the smoke test, and applying migration 196 is the only remaining step before this is live. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
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
1 participant
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.
Branch for review/comparison against
main. Everything here is proposed and CI-gated — the TS/RN changes are hand-verified only (no local build/tests were possible); Python services passpy_compile. Please let CI + review validate before merge.What's in it
docs/qa/— full QA audit of the app (TestFlight coach + client walkthrough + code audit):FITFO_BUGS.md— 68 findings withfile:lineFITFO_BUGS_STATUS_ON_MAIN.md— re-validation vs shippedmain(12 already fixed, 3 fixed-in-effect, 43 open)FITFO_FLOWS.md,FITFO_SUBSYSTEMS.md— flows to redesign + how the engine worksFITFO_TIER_30_COMMUNITY_SPEC.md— buildable spec for a new $30 community tier (real backend-wired plan, not the client-side demo mock)23 safe mechanical fixes (engine intentionally untouched — 2 working sets + goal→reps are product decisions):
linkedspread fix, redundant plan modal removal, leaderboard real counts, equipment prefill contradiction, recent-workout dedupe, A/B leg labels, nutrition load race, hub poll gating, check-in copy.Not included (need a decision, not a blind edit)
Engine rep/volume (B02/B03), leaderboard coach-exclusion SQL migration (B21), adherence table wiring (B15), plan-state enum refactor (B11/B43), client dedupe/count definitions (B12/B13), rank off-by-one (B24), overdue check-in UX (B26/B27), billing/referral policy (B50/B51), canonical set-count (B04), prompt-injection/window (B40/B42).
Notes
origin/main(the branch build 102 shipped from); the QA line numbers inFITFO_BUGS.mdwere from an older branch, so the status doc has the realmainlocations.TourPage.tsxwas caught and excluded.🤖 Generated with Claude Code