Test author
Guardrails, skills, loops, hooks & review agents for database + website builders on Claude Code (Next.js + Supabase).
npx -y skills add m-binimran/dev-pack --skill test-authorAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Write tests first (RED→GREEN→REFACTOR) with Vitest, Testing Library, and Playwright for a Next.js + Supabase app. Use when adding a feature, fixing a bug, or raising coverage.
SKILL.md
1.6 KB, as published. Nobody here has run it
test-author
Tests that prove behavior and catch regressions — written before or alongside the code, then actually run.
Process
- Start RED. Write a failing test that states the expected behavior. Run it; confirm it fails for the right reason.
- GREEN. Write the minimum code to pass. Run it.
- REFACTOR. Clean up with the test as a safety net. Re-run.
What to write where
- Unit (Vitest): pure functions, Zod schemas, utils. Cover edge cases + error paths, not just happy path.
- Component (Vitest + Testing Library): query by role/label (also exercises a11y), simulate user events, assert visible outcomes — not internal state.
- E2E (Playwright): the critical flows only (auth, core task, checkout). Keep these few and stable.
Patterns
- Bug fix → regression test that fails before the fix, passes after. Reference the bug.
- Deterministic data: fixed seeds/fixtures (
seed-fixtures); no real network/clock in unit tests — mock them. - Arrange-Act-Assert, one behavior per test, names that read as sentences.
Output
- The test file(s), and the real run output (pass/fail counts). A test you didn't run is unverified — say so.
Guardrails
- Don't test implementation details that change on every refactor.
- Don't lower coverage thresholds or skip tests to go green — fix the code or the test.