Bring docs in line with the post-AM-6 state so a fresh agent can pick
up cleanly without first reconciling stale numbers:
- pytest count: 88/59/31 → 92 across all references
- live scrape: 9,960 rows / 36s → 9,980 rows / 44s (latest run, 2026-05-06)
- migrations applied: 0001-0005 → 0001-0007 in both docs
- verification gate updated with auto-minted JWT detail and the 29,779
ingredient_grocery_match rows produced post-scrape
- "What is real" / Backend: added swiftly_auth.py bullet describing
get_token() / mint_anonymous_token() / cache semantics + the
10,928-item live verification
- file map: added services/swiftly_auth.py, services/matcher.py,
services/planner/, mentioned scraper_service runs matcher post-scrape
- file map: alembic versions 0001 → 0007, tests/ count = 92, marked
config.py as no-longer-carrying SWIFTLY_BEARER_TOKEN
- spec map line for swiftly-token-auto-mint: "next-up" → "Implemented"
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Breaks the auto-mint spec into 6 ordered, sized tasks with an explicit
halt-for-approval boundary at AM-2 (live scrape verification before
removing the env var). Restates the verified prerequisites (config.json
publicly readable; Firebase signUp returns valid JWT with proper headers;
Swiftly accepts the minted token) so a fresh agent doesn't have to
re-discover them.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Discovery: luckysupermarkets.com/config.json is publicly readable and
exposes firebaseApiKey. With proper Origin/Referer headers, Firebase
Identity Toolkit's anonymous-signup REST endpoint mints the same JWT
shape (iss=swiftly-lu-prod, aud=swiftly-lu-prod, anon provider, 3600s
TTL) that Swiftly accepts. Verified end-to-end on 2026-05-06.
This eliminates the manual hourly token-capture toil and supersedes
the seleniumbase-based scripts/refresh_swiftly_token.py (commit
ccfb38a) which had partial UI selector issues.
- New spec: docs/specs/2026-05-06-swiftly-token-auto-mint.md
- HANDOFF.md TL;DR refreshed (Phase 9 shipped); caveat #2 + #3
rewritten to point to the auto-mint redesign; suggested-next-move
reordered to put the redesign first
- ORIENTATION.md env-var section flags SWIFTLY_BEARER_TOKEN as
scheduled-for-removal; "Where to look" lists both specs;
last-updated footer refreshed
Implementation deferred — this commit captures the design and routing
only. Estimated 2-3 hours of focused work to ship per the spec.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
8-task TDD plan covering planner package (config, types, cost,
filter, score, select, generate orchestrator), generate/regenerate/
get endpoints, and end-to-end smoke test against seeded data.
Builds on the thin phase 4 plan; uses NeverSuggest for blocklists,
ingredient_grocery_match for sale-aware cost estimation, top-K=20
set enumeration with diversity penalty for protein/cuisine repeats.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
15-task TDD plan covering migration 0006 (ingredient.aliases,
recipe.calories_per_serving, ingredient_grocery_match), CRUD endpoints
for ingredient/recipe/never-suggest, rapidfuzz matcher service hooked
into scrape success, manual match override, resolve-ingredient assist,
and migration 0007 seeding 30 starter recipes.
Reuses existing NeverSuggest table for ingredient + recipe blocking
rather than adding a new family_ingredient_block table.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Captures brainstorm decisions for the meal-planner algorithm and the
minimum recipe-engine surface needed to feed it:
- 6 hard constraints (blocklist, never_suggest, recency N=4, calories
+/-20%, time <=45min, cost <=$30)
- Top-K=20 set enumeration with diversity penalty for protein/cuisine
- Ingredient<->grocery_item matching as a cacheable layer (rapidfuzz +
manual override) rather than per-run fuzzy work
- Thin phase 4: recipe CRUD, ingredient CRUD, resolve-ingredient
assist, manual match override, match job, 30-recipe seed
- Recipe ingestion source pros/cons (TheMealDB + Spoonacular + manual)
documented; decision deferred until phase 4 + phase 9 work end-to-end
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Updates ORIENTATION.md and HANDOFF.md to reflect actual state as of
commit 8e89f79: phases 1/2/3/7 complete (verified, not just claimed),
phases 4/5/6/8/9/10/11 not started. Documents the auth model, the
bootstrap login hatch, the SWIFTLY_BEARER_TOKEN expiration handling,
the Swiftly JSON API ingestion path that replaced Playwright, the
canonicalized API paths, and the verification commands to reproduce
the 31/31 pytest gate locally and in CI.
HANDOFF.md is intended for fresh agents and points at .agent/
phase-summaries for the per-phase write-ups. Surfaces 10 caveats and
traps the next agent will hit if they skim, and recommends Phase 9
(meal-planner generation algorithm) as the next move.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
adminandClaude Opus 4.7