playwright-cli

Run playwright-cli open with a URL, then playwright-cli snapshot to get element refs like e5. Use commands such as click, fill, and type with those refs, and close the browser with playwright-cli close when finished.

Run the test with npx playwright test --debug=cli in the background and wait for the printed session name. Attach with playwright-cli attach tw-XXXX, then use snapshot, console, and requests to diagnose the failure and rehearse the fix.

Yes, use playwright-cli attach --cdp=chrome or --cdp=msedge after enabling remote debugging in the browser via chrome://inspect. You can also attach to any CDP endpoint URL or through the Playwright browser extension.

Use playwright-cli route with a URL pattern and options like --status, --body, or --header to intercept requests. For conditional logic or response modification, use run-code with page.route in a custom function.

Yes, the -s flag creates named sessions with independent cookies, storage, cache, and tabs. Use playwright-cli list to see sessions, close to stop one, and close-all or kill-all for cleanup.

Recording and tracing add overhead to automation and can consume significant disk space. Traces capture DOM snapshots and network detail for debugging, while video produces WebM files better suited for demos and documentation.