Playwright: Let tests, scripts or AI agents click through websites for you

What it does

Playwright drives the Chromium, Firefox and WebKit browser engines through one API. It comes in several forms: Playwright Test, a test runner for end to end testing (checking a whole website the way a user would); a library for scripts; a CLI designed for coding agents; and Playwright MCP (a way to plug tools into an AI assistant). With MCP, the agent reads the page as an accessibility tree, a structured list of buttons and fields, so it does not need screenshots or vision models to click and type.

How to use it

Input → action → output: write a test or script that says which page to open, what to click or type, and what should appear, then run it. You get pass or fail results, and on failure you can capture traces, screenshots and videos to inspect each step. With MCP, you instead ask your AI assistant to do something on a web page and it performs the clicks.

Common use cases

• Write a test that logs into your web app and checks the dashboard loads, then run it in three browser engines in parallel.
• Connect Playwright MCP to your AI assistant and ask it to fill out a web form; it reads the page and types into each field.
• Write a short script that visits a list of pages and saves a screenshot or PDF of each one.