review-test — audit test quality on touched files

review-test doesn't ask "is there a test"; lots of repos have lots of tests that don't catch bugs. It asks:

• Is the test at the right level? Unit / integration / e2e.
• Does it actually assert the behaviour under test?
• Does it mock something that shouldn't be mocked?
• Will it fail when the behaviour breaks, and only then?
• Is it independent — runs alone, in any order, in parallel?

Flags missing coverage on newly added code paths, runs the suite to surface failures, emits findings in the finding-format.md schema. Never edits files. To act on the findings, call /drive-change (which dispatches fix-applier agents under sensitivity gating).

Phase 0 — Scope

In priority order:

1. PR context: gh pr diff --name-only against the base branch.
2. Working tree: git diff --name-only HEAD.
3. Explicit list from the user.