Browser-Use enables AI agents to access web-based content through non-headless browsing sessions, allowing them to extract data and perform actions that would otherwise be unavailable.

​
Key Capabilities

Web Data Extraction

Extract structured data from websites even when no API is available

Browser Automation

Automate web interactions to collect data or perform tasks

Real Browser Rendering

Access JavaScript-rendered content and dynamic websites

Session Logging

Track browser sessions with detailed logs for transparency

​
Current Limitations

  • Does not yet support authenticated browsing sessions
  • Cookies and session data are not persisted between requests
  • Limited to publicly accessible websites

​
Requirements

1

API Key

Generate and use an MCP API key with appropriate permissions

​
Endpoint

https://api-gateway.cobrowser.xyz/mcp/browser/sse?api_key=YOUR_API_KEY

​
Implementation with Python SDK

import asyncio
import logging

from mcp import ClientSession
from mcp.client.sse import sse_client

logging.basicConfig(
    level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s"
)

async def main():
    async with sse_client(
        url="https://api-gateway.cobrowser.xyz/mcp/browser/sse?api_key=YOUR_API_KEY",
    ) as streams:
        async with ClientSession(*streams) as session:
            await session.initialize()

            # List available tools
            tools = await session.list_tools()
            print(tools)

            # Browser navigation example
            result = await session.call_tool(
                "browser_use",
                {
                    "url": "https://example.com",
                    "action": "navigate"
                }
            )
            print(result)

asyncio.run(main())

​
Implementation in Cursor

To use Browser-Use in Cursor, create an mcp.json file with the following configuration:

{
  "mcpServers": {
    "browser-use-mcp-server": {
      "url": "https://api-gateway.cobrowser.xyz/mcp/browser/sse?api_key=YOUR_API_KEY"
    }
  }
}

Place this file in the .cursor folder in your project directory to enable browser capabilities for your AI agents within Cursor.

​
Use Cases for AI Agents

  • Retrieve information from websites that do not offer APIs
  • Collect data from multiple sources for analysis
  • Process visual information on web pages
  • Read content from dynamically generated web pages

Next: Email-Use for Gmail Integration