Files
ai-for-dummies/plans/astro-refactor/task-06-content-review.md
T

46 lines
1.7 KiB
Markdown

# Task 06 — Review-desk content out of catalog.js
**Agent**: `content-i18n-migrator` · **Model**: MiniMax-M3 **Depends on**: 04 ·
**Parallel with**: 05 · **Blocks**: 16 **Worktree**:
`.agents/scripts/worktree.sh start 06 content-review`
## Goal
24 review entries become a typed collection, with the `improved` markdown as
real `.md` files.
## Scope
`src/content/reviews/**`, `scripts/build-skill-review.mjs`.
## Steps
1. Move all 24 entries from `skills-review/catalog.js` (27 KB) and
`submitted-catalog.js` (18 KB) into the collection.
2. `improved` becomes a real Markdown file per entry, rendered at build time.
That deletes the hand-rolled client-side renderer.
3. **Re-point `scripts/build-skill-review.mjs`** at the collection.
## The two traps
1. **The generator writes committed files.**
`skill-reviews/improved/**/SKILL.md` is generated from the `improved` fields
and the output is in git. Move `catalog.js` and the generator keeps running
against nothing — **silently, exit code 0**. Prove it still produces
byte-identical output:
```bash
node scripts/build-skill-review.mjs && git diff --exit-code skill-reviews/
```
2. **The diff view needs raw source.** The review desk compares original and
improved as _text_. If `improved` only exists as rendered HTML, the change
lens breaks. Keep the raw string reachable.
## Done when
- [ ] 24 entries in the collection; `verify.mjs`'s `id:'` count assertion still
passes
- [ ] `git diff --exit-code skill-reviews/` clean after regenerating
- [ ] `astro check` passes
- [ ] Every reference to the generator still accurate (`package.json`,
`README.md`, `docs/operations-guide.md`, the review desk footer)