Agent-MCP-Browser-Use

efengx/Agent-MCP-Browser-Use

3.1

If you are the rightful owner of Agent-MCP-Browser-Use and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The Agent MCP Browser Use server provides AI with the capability to autonomously interact with the web, converting language instructions into live web actions.

Agent Mcp Browser Use

Our vision is to give AI "hands and eyes" on the web. This server acts as the core, taking instructions from any AI to automate a browser. It enables autonomous interaction for everything from simple queries to complex operations, seamlessly converting language into live web actions.

Features

🤖 AI-Powered Browser Control: Empower your AI agents to navigate, click, type, and scrape any website, turning abstract goals into concrete actions.

🔌 Flexible Transport: Integrates seamlessly with AI clients via both SSE (Server-Sent Events) and stdio, ensuring broad compatibility.

👀 Real-Time VNC Viewing: Watch your agent browse the web in real-time, just like screen sharing. Perfect for debugging, demos, and seeing your AI work its magic.

⚡️ Asynchronous by Design: Handles complex, long-running browser tasks without blocking, ensuring a responsive and efficient experience.

Environment

Create a .env file in your project root:

OPENAI_API_KEY=your-api-key
CHROME_PATH=optional/path/to/chrome
PATIENT=false  # Set to true if API calls should wait for task completion

Installation

# Install dependencies from pyproject.toml
uv sync
# Install Playwright and its browser dependencies
uv pip install playwright
uv run playwright install --with-deps --no-shell chromium

Run the server directly from the source for quick testing.

# Run directly from source
uv run server --port 8000

Build and install the package globally for use in production or with stdio-based clients.

# 1. Build and install globally
uv build
uv tool uninstall browser-use-mcp-server 2>/dev/null || true
uv tool install dist/browser_use_mcp_server-*.whl

# 2. Run with stdio transport
browser-use-mcp-server run server --port 8000 --stdio --proxy-port 9000

Configure your AI agent client to connect to the server.

SSE Mode Client Configuration

{
  "mcpServers": {
    "browser-use-mcp-server": {
      "url": "http://localhost:8000/sse"
    }
  }
}

{
  "mcpServers": {
    "browser-server": {
      "command": "browser-use-mcp-server",
      "args": [
        "run",
        "server",
        "--port",
        "8000",
        "--stdio",
        "--proxy-port",
        "9000"
      ],
      "env": {
        "OPENAI_API_KEY": "your-api-key"
      }
    }
  }
}