[#185] DB schema: title column on plots + comments table + API

[realproject7]

Summary

• Migration: title TEXT DEFAULT '' on plots table; comments table with indexes on (storyline_id, plot_index) and (commenter_address), RLS for public read
• Types: Added title to Plot type, added Comment type alias
• Plot indexer: Extracts title from updated PlotChained event and stores it
• Comments API: GET /api/comments?storylineId=N&plotIndex=M (paginated, 20/page) + POST /api/comments (wallet signature auth, 1 comment per address per plot per minute, max 1000 chars)

Test plan

• Run migration on Supabase
• Chain a plot with title → verify title stored in plots table
• Chain a plot without title → verify empty string stored
• POST /api/comments with valid signature → success
• POST /api/comments twice within 1 min → 429
• GET /api/comments returns paginated results
• Hidden comments not returned by GET

Fixes #185

🤖 Generated with Claude Code

[project7-interns]

Verdict: REQUEST CHANGES

Summary

The schema and indexer pieces look aligned with #185, but the new comments read API does not match the API contract defined in the issue. That makes the foundation inconsistent for the blocked Phase 9 comment UI work.

Findings

• [high] GET /api/comments implements the wrong pagination contract
  • File: src/app/api/comments/route.ts:14
  • File: src/app/api/comments/route.ts:30
  • Suggestion: implement the issue-specified page and limit query params, or at minimum accept them compatibly and derive the Supabase range from those values. Right now the route only accepts offset and hardcodes PAGE_SIZE = 20, but issue #185 explicitly defines GET /api/comments?storylineId=X&plotIndex=Y&page=1&limit=20.

Decision

Requesting changes because #185 is the schema/API foundation for later comment UI work, and the shipped API contract should match the issue rather than introducing a different pagination shape.

[project7-interns]

Verdict: APPROVE

Summary

The follow-up change fixes the API contract mismatch by implementing page and limit query params on GET /api/comments, which aligns the route with issue #185. The schema, types, plot-title indexing, and comments API foundation now look consistent for the blocked Phase 9 frontend work.

Findings

• None.

Decision

Approving because the previously blocking pagination contract gap is resolved, and GitHub lint-and-typecheck passes on the updated head.

[project7-interns]

T2b: APPROVE

Well-structured schema + API foundation for Phase 9:

• Migration: title on plots, comments table with proper indexes, RLS filtering hidden rows
• Plot indexer: Correctly extracts title from updated PlotChained event
• Comments GET: page/limit pagination with total count — matches issue #185 spec
• Comments POST: Strong auth — wallet signature binds to specific comment content (prevents replay), 1/min rate limit per address per plot, 1000 char max
• Types: Supabase types and Comment export updated consistently

Minor nit: GET route code comment (line 16) still references offset=0 but code uses page/limit. Cosmetic only.