playwright-cli

Use playwright-cli to open a browser, navigate to a URL, and interact with elements using refs from the accessibility snapshot. Commands like click, fill, and type target elements such as e5, and each command returns an updated page snapshot.

Run the test with npx playwright test --debug=cli in the background, then attach with playwright-cli attach using the printed session name. The paused page can be explored with snapshots, console logs, and network requests to find the cause.

Yes, every playwright-cli action prints the equivalent Playwright TypeScript code, such as page.getByRole locators. You collect these generated lines into a test file and add assertions manually using matchers like toBeVisible or toHaveText.

Yes, the -s flag creates named sessions with independent cookies, localStorage, cache, and tabs. Sessions can run concurrently, use persistent profiles, and be listed, closed, or cleaned up with list, close-all, and delete-data commands.

Use playwright-cli route with a URL pattern to return custom status codes, JSON bodies, or headers, and unroute to remove rules. For conditional logic or response modification, run-code executes arbitrary Playwright routing functions.

Video recording produces WebM files and adds slight overhead to automation, with large recordings consuming significant disk space. For debugging with DOM snapshots and network details, tracing is more appropriate than video.