MCP Servers / productivity / Browser Use MCP Server

Browser Use MCP Server

JSON →

none★ 3productivity

Automate browser actions using natural language commands. Powered by Playwright and supports multiple LLM providers.

github ↗

Install

How to run this server

[ { "cmd": "pip install -e", "imports": [] } ]

server path: browser-use-mcp-server-2

Tools

What this server exposes

browse_web

Navigate to a URL and retrieve the page content

search_web

Search the web using a search engine and return results

click_element

Click on a specific element on the page identified by selector or text

fill_form

Fill out a form field with specified text

extract_content

Extract specific content from the current page

take_screenshot

Capture a screenshot of the current browser viewport

scroll_page

Scroll the page in a specified direction

execute_javascript

Execute custom JavaScript code in the browser context

get_page_info

Get information about the current page such as URL, title, and metadata

navigate_back

Navigate back to the previous page in browser history

navigate_forward

Navigate forward to the next page in browser history

refresh_page

Refresh the current page

close_browser

Close the browser session

Configuration

Environment & auth

authnone

envOPENAI_API_KEY

envANTHROPIC_API_KEY

envGOOGLE_API_KEY

envCOHERE_API_KEY

envMISTRAL_API_KEY

envGROQ_API_KEY

envTOGETHER_API_KEY

envAWS_ACCESS_KEY_ID

envAWS_SECRET_ACCESS_KEY

envFIREWORKS_API_KEY

envAZURE_OPENAI_API_KEY

envNVIDIA_API_KEY

envAI21_API_KEY

envDATABRICKS_HOST

envDATABRICKS_TOKEN

envWATSONX_API_KEY

envXAI_API_KEY

envUPSTAGE_API_KEY

envHUGGINGFACE_API_KEY

envOLLAMA_BASE_URL

Resources

githubhttps://github.com/pietrozullo/browser-use-mcp ↗