review

This skill systematically reviews Playwright test files for quality problems, checking them against a curated catalogue of twenty anti-patterns spanning flakiness, maintainability, and coverage gaps. It solves a common problem in end-to-end testing: suites that pass locally but flake in CI because of arbitrary waits, non-web-first assertions, hardcoded URLs, brittle CSS/XPath selectors, missing awaits, and order-dependent shared state. Instead of ad-hoc code review, it gives the agent a repeatable rubric to grade tests objectively.

The workflow gathers context from `playwright.config.ts` and the target spec files, then checks each file against critical, warning, and informational anti-patterns drawn from the bundled `anti-patterns.md` reference. It scores each file on a 1-10 scale, produces a structured per-file report with line-level findings and suggested replacements (for example swapping `waitForTimeout` for a web-first `expect().toBeVisible()`), and can offer to apply fixes with the Edit tool. For large suites it can spawn parallel sub-agents per file and aggregate results into a summary table, also surfacing coverage gaps where pages or features have no tests.

It targets QA engineers, SDETs, and developers maintaining Playwright test suites who want to reduce flakiness and enforce best practices before merging. Typical use cases include auditing an inherited test suite, gating pull requests on test quality, and coaching teams toward role-based locators and independent, well-named tests.