Playwright
OFFICIALGitHub
Playwright-mcp gives LLMs access to browser automation.
Most interesting: It uses accessibility features, so the LLM doesn't need to try and parse screenshots.
From the Readme:
Snapshot Mode
browser_navigate
- Navigate to a URLbrowser_go_back
- Go back to the previous pagebrowser_go_forward
- Go forward to the next pagebrowser_click
- Perform click on a web pagebrowser_hover
- Hover over element on pagebrowser_drag
- Perform drag and drop between two elementsbrowser_type
- Type text into editable elementbrowser_press_key
- Press a key on the keyboardbrowser_snapshot
- Capture accessibility snapshot of the current page (better than screenshot)browser_save_as_pdf
- Save page as PDFbrowser_wait
- Wait for a specified time in secondsbrowser_close
- Close the page
Vision Mode
browser_navigate
- Navigate to a URLbrowser_go_back
- Go back to the previous pagebrowser_go_forward
- Go forward to the next pagebrowser_screenshot
- Capture screenshot of the current pagebrowser_move_mouse
- Move mouse to specified coordinatesbrowser_click
- Click at specified coordinatesbrowser_drag
- Perform drag and drop operationbrowser_type
- Type text at specified coordinatesbrowser_press_key
- Press a key on the keyboardbrowser_save_as_pdf
- Save page as PDFbrowser_wait
- Wait for a specified time in secondsbrowser_close
- Close the page