What problem does it solve? Manually driving a browser to test web pages, debug failing Playwright tests, or capture UI evidence is slow and repetitive. This Skill lets an agent open browsers, click and fill elements via snapshot refs, mock network requests, manage storage state, and generate Playwright test code directly from terminal commands. ## Core Features & Use Cases - Browser Automation: Open pages, click, fill, drag, and navigate using element refs from accessibility snapshots, with support for tabs, sessions, and persistent profiles. - Test Authoring and Healing: Run Playwright tests with --debug=cli, attach to the paused session, and turn every CLI action into generated TypeScript test code following a plan-generate-heal workflow. - Debugging and Evidence Capture: Record traces, videos with chapter overlays, screenshots, console logs, and network requests, then attach them to pull requests with gh --attach. - Use Case: A failing CI test needs investigation. Run it with --debug=cli, attach playwright-cli to the paused session, inspect the snapshot and console, rehearse the fix interactively, and paste the generated locator code back into the test file. ## Quick Start Ask the agent to open a website with playwright-cli, take a snapshot, and click a specific element by its ref.