Everything here needs a page that has actually rendered: a screenshot of what a user sees, a form walked to the end, a value that only exists after JavaScript runs. Nothing is installed locally to get it.
A browser automation server backed by BrowserCat's hosted browsers. Seven tools cover navigation, clicking, hovering, filling inputs, selecting from dropdowns, screenshots and JavaScript execution. Console output and captured screenshots are also exposed as resources at console://logs and screenshot://name.
- Any URL opened in a cloud browser
- A screenshot of the full page or of one element, at a width and height you choose
- Clicks and hovers on a CSS selector
- Input fields filled and dropdown options selected by value
- JavaScript executed in the page, with its result returned
- The browser's console output, readable as a resource
A BrowserCat API key in BROWSERCAT_API_KEY; free keys are issued at browsercat.xyz/mcp. Launched with npx against @browsercatco/mcp-server. No local browser install — the browser runs on BrowserCat's side, so the pages you visit are fetched from their infrastructure.
One command plus a key — npx -y @browsercatco/mcp-server, then supply credentials