This MCP server enables AI to control web browsers for tasks such as web scraping and form filling. It integrates with various AI models to automate browser actions.
From the registry: AI browser automation - navigate, click, type, extract content, and run autonomous web tasks
$ uvx playwright install chromium --with-deps --no-shellPlease install the `browser-use` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install Chromium** — Install Playwright Chromium needed by browser-use Run: `uvx playwright install chromium --with-deps --no-shell`
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["browser-use[cli]","--mcp"]
- optional environment variables:
- `OPENAI_API_KEY`: OpenAI API key for the agent's LLM (example: `sk-<placeholder>`)
- `GEMINI_API_KEY`: Gemini API key for the agent's LLM (example: `<placeholder>`)
- `BROWSER_USE_API_KEY`: Browser Use cloud API key for stealth cloud browsers (example: `<placeholder>`)
Note: At least one LLM provider key (OpenAI/Gemini/etc.) is required.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.GEMINI_API_KEYrequiredAPI key for accessing the Gemini AI model.BROWSER_USE_API_KEYrequiredAPI key for accessing the Browser-Use cloud service.OPENAI_API_KEYrequiredAPI key for accessing OpenAI services.MCP server for Firecrawl web scraping, structured data extraction and web search integration.