coverage

This skill analyzes test-coverage gaps in an application, focused on end-to-end/Playwright testing. It solves the problem of not knowing what is actually tested by mapping all testable surfaces (routes, components, API endpoints, and critical user flows) and comparing them against the tests that already exist, so teams can see where their coverage is thin.

The workflow uses an Explore subagent to catalog routes and pages (Next.js app directory, React Router, Vue Router, etc.), interactive components, and API endpoints, then scans existing `*.spec.ts`/`*.spec.js` files to determine which pages, components, and endpoints are exercised. It produces a coverage matrix table with covered/partial/missing status, prioritizes uncovered areas by business impact (critical, high, medium, low), and suggests a concrete test plan with the number of tests needed, a recommended template, and an effort estimate. It can optionally auto-generate tests for the top gaps by invoking a companion `/pw:generate` command.

It targets QA engineers and developers using Playwright who want a structured, prioritized view of coverage rather than a raw percentage. It reads source and test files and can trigger test generation, but its own analysis is read-only inspection of the codebase and produces reports and recommendations.