What problem does it solve? Browser tests break when selectors go stale, routes change, or timing becomes flaky, and fixing them by hand requires tedious reproduction and root-cause analysis. This Skill diagnoses one failing test file, applies a minimal repair, and verifies it passes. ## Core Features & Use Cases - Failure Diagnosis: Reproduces the failing test in isolation and classifies the root cause, such as stale selectors, changed routes, timing issues, or real application regressions. - Minimal Test Repair: Updates selectors, test helpers, or test config while preserving the intended assertion, preferring resilient role, label, and text selectors. - Feature Plan Sync: Marks the relevant criterion complete in docs/features browser-test plans only after the fixed test passes. - Use Case: A Playwright test that passed last week now fails after a UI rename. Point the Skill at the failing file, and it reproduces the failure, updates the selector, re-runs the test, and reports the root cause. ## Quick Start Fix the failing Playwright test in tests/checkout.spec.ts and verify it passes.
