Files

T

Marcos Paulo 48c31dc1b3 build: migrate from npm to pnpm

Ten git worktrees each carried their own 225 MB node_modules (1.1 GB across
five) and paid 11s per `npm ci`. pnpm hardlinks from a shared store: the same
five worktrees cost ~250 MB total, and a fresh install is 4s.

What changed beyond the mechanical rename:

- `overrides` moved to `pnpm-workspace.yaml`. pnpm 11 does not read the `pnpm`
  field in package.json *or* npm's top-level `overrides`, and it fails silently
  — the vite/defu/language-server pins would have quietly stopped applying.
- Build scripts are blocked by default in pnpm; esbuild and sharp are allowed
  explicitly via `allowBuilds` (renamed from `onlyBuiltDependencies` in 11).
- `packageManager` + `engines` pin the toolchain.
- gate.sh rejects a package-lock.json/yarn.lock/bun.lock outright, so an agent
  running `npm install` out of habit fails loudly instead of building a second,
  divergent dependency tree.
- CI bootstraps pnpm with `npm install --global pnpm@11.25.0` rather than
  corepack (unbundled as of Node 25) or pnpm/action-setup (this self-hosted
  act-runner has never run a job; fetching a third-party action is not
  something to discover on the first one).

Two pre-existing CI bugs fixed while in the file:

- the gate installed with `npm install --package-lock=false`, which discarded
  the lockfile the previous session had just fixed.
- the visual-regression step imported `playwright`, which is not a dependency,
  and `visual-regression.mjs` has no compare mode anyway — in CI it overwrote
  its own baselines and passed unconditionally. Removed with a comment; it
  comes back when it can diff.

The `publish` job is now manual (`workflow_dispatch`). During the migration
dist/ holds three HTML files against the live pages branch's ten, so publishing
on every push to main would take the site down to a stub. Restore at task 20.

HANDOVER.md's incident log still says npm where it describes what happened at
the time; that is history, not a missed rename.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

2026-09-05 04:29:42 +00:00

1.8 KiB

Raw Blame History

Task 18 — Motion pass

Agent: motion-designer · Model: Gemini (perceptual judgement) Depends on: 15, 16 · Parallel with: 19 Worktree: .agents/scripts/worktree.sh start 18 motion

Goal

Audit the motion that exists, add only motion that earns its place, and make prefers-reduced-motion correct everywhere.

Order of work — audit first

The audit is worth more than the additions.

Inventory every transition, animation, @keyframes, transform in src/.
Flag anything animating a layout property (width, height, top, left, margin). Those force reflow every frame and fail the 200ms INP budget. The Stylelint config already blocks new ones; find the ported ones.
Confirm every animated element has a reduced-motion path, then test it under DevTools emulation. Reduced means reduced, not broken — end states must still be correct.

Then, sparingly

Candidates, each needing a written purpose:

tab panel change in the guide (state change — justified)
review-desk detail swap on selection (state change — justified)
vote widget tally update (feedback — justified)

transform and opacity only. 150–250ms. cubic-bezier(.2,0,0,1). One thing moves at a time — no staggered card cascades; this design has a point of view and staggering reads as a template.

View transitions — optional, decide explicitly

Astro's <ClientRouter /> is the only sanctioned motion dependency. If adopted, verify: JS-disabled navigation, browser back/forward, the review desk's six query params, and reduced motion. If any fails, do not ship it.

Done when

Zero animations on layout properties
Every animation has a one-line written purpose
Reduced-motion tested under emulation; end states correct
Before/after captures attached; pnpm run gate green

1.8 KiB Raw Blame History Unescape Escape