🔗 Official Repository
github.com/kontext-dev/browser-use-mcp-serverAlways check the official repository for the latest setup instructions and license.
What is it?
browser-use packaged as an MCP server with SSE transport. includes a dockerfile to run chromium in docker + a vnc server. — is a Model Context Protocol (MCP) server enabling AI applications like Claude and Cursor to securely access browser automation capabilities through the open protocol.
Why kontext-dev/browser-use-mcp-server?
browser-use-mcp-server is a Model Context Protocol (MCP) server that empowers AI agents to automate and control web browsers using browser-use and Playwright. It supports real-time interaction monitoring via VNC streaming, asynchronous task execution, and flexible connection through both SSE and stdio transport modes.
Common Use Cases
- Enabling AI agents to navigate and interact with web pages automatically
- Watching real-time browser automation via VNC streaming during task execution
- Integrating browser automation into MCP clients like Cursor, Windsurf, and Claude Desktop
Quick Install (Claude Desktop)
{
"mcpServers": {
"browser-use-mcp-server": {
"command": "npx",
"args": ["-y", "kontext-dev/browser-use-mcp-server"]
}
}
}Add to your Claude Desktop config file after installing requirements per the repo instructions.
Frequently Asked Questions
What prerequisites are required to run browser-use-mcp-server?
You need the uv package manager, Playwright with Chromium installed, and mcp-proxy if you intend to run in stdio mode.
Which transport protocols are supported by the server?
The server supports both Server-Sent Events (SSE) mode and standard input/output (stdio) mode.
Can I watch the browser automation live as it executes?
Yes, the server features built-in VNC streaming, allowing you to view browser automation in real-time.