Playwright
Browser testing that an agent can write and run itself.
Drives Chromium, Firefox, and WebKit with automatic waiting that removes most timing flakiness. In agentic workflows it does double duty: the test suite is both your safety net and the feedback signal the agent uses to know whether it's done.
Best for
Proving the app still works after the AI changed ten files.
Watch out for
Flaky selectors produce false alarms. Prefer roles and visible text over CSS paths.
Using this and stuck anyway?
Knowing the tool isn't the same as finishing the project. If yours has stopped moving, we'll take a look at it.
Tell us what's broken