What problem does it solve? React test suites often drift into brittle patterns: overuse of getByTestId, fireEvent instead of userEvent, oversized snapshots, and assertions against raw utility classes. This Skill statically audits a TypeScript and React project's tests against an opinionated four-layer baseline and produces structured findings plus an optional implementation plan. ## Core Features & Use Cases - Four-layer audit: Covers test runner and tooling, query priority and selector hygiene, interaction and async patterns, and test design and coverage, each graded as present, partial, missing, or violation. - Diagnostic snapshot: Measures query distribution across the Testing Library priority ladder, userEvent vs fireEvent ratios, snapshot sizes, and flaky-pattern signals, written to .architect-audits/testing-audit/. - Optional coverage enrichment: With --with-run, invokes Vitest or Jest in coverage mode and folds real coverage data into the findings; never runs Playwright or Cypress. - Use Case: Run the audit on a React codebase before a refactor to find tests coupled to implementation details, then accept the generated implementation plan that orders fixes by testing-philosophy impact. ## Quick Start Ask the assistant to run /testing-audit on this repository to audit the React tests and report the top findings.
