browser-use MCP Server

What does “browser-use” MCP Server do?

The browser-use MCP (Model Context Protocol) Server enables AI agents to control web browsers programmatically using the browser-use library. This server acts as a bridge between AI assistants and web browsers, allowing automated browsing, web data extraction, and interaction with websites directly from development environments such as Cursor. By exposing browser automation capabilities to AI agents, it streamlines workflows like searching the web, scraping content, filling forms, and navigating sites, all under programmatic control. This enhances development by automating repetitive web tasks and providing real-time web context to AI assistants.

List of Prompts

No prompt templates are mentioned or documented in the repository.

List of Resources

No explicit resources are documented or listed in the repository.

List of Tools

Tools are not explicitly documented in the root or main README, and server.py is not directly exposed in the structure. No detailed tool list is available from public documentation.

Use Cases of this MCP Server

• Automated Web Browsing: Enable AI agents to open websites, follow links, and interact with page elements for research or information gathering.
• Web Data Extraction: Automatically scrape text, tables, and other content from web pages to provide context or data for further processing.
• Form Interaction and Submission: Programmatically fill out and submit forms (e.g., login, search) as part of automated workflows.
• Test Automation: Use AI agents to perform browser-based testing of web applications, ensuring UI flows work as expected.
• Live Web Context for AI: Allow AI assistants to retrieve real-time information (e.g., latest news, stock prices) by browsing the web on demand.

How to set it up

Windsurf

1. Ensure you have Node.js and the required environment for Windsurf.
2. Open your Windsurf configuration file (e.g., windsurf.config.json).
3. Add the browser-use MCP Server to the mcpServers section:

   {
     "mcpServers": {
       "browser-use": {
         "command": "npx",
         "args": ["@browser-use/mcp-server@latest"]
       }
     }
   }
   

4. Save the file and restart Windsurf.
5. Verify the setup by checking for the MCP server in the Windsurf interface.

Claude

1. Ensure Claude supports external MCP servers and you have Node.js installed.
2. Edit the Claude config file to add the MCP server:

   {
     "mcpServers": {
       "browser-use": {
         "command": "npx",
         "args": ["@browser-use/mcp-server@latest"]
       }
     }
   }
   

3. Save, restart Claude, and confirm integration by running a test action.

Cursor

1. Verify prerequisites: uv, Playwright.
2. Open Cursor’s configuration (e.g., .cursor/config.json).
3. Add the MCP server:

   {
     "mcpServers": {
       "browser-use": {
         "command": "npx",
         "args": ["@browser-use/mcp-server@latest"]
       }
     }
   }
   

4. Save and restart Cursor.
5. Test by triggering a browser automation action from Cursor.

Cline

1. Make sure Cline is installed and configured for MCP servers.
2. Edit the Cline config to register the MCP server:

   {
     "mcpServers": {
       "browser-use": {
         "command": "npx",
         "args": ["@browser-use/mcp-server@latest"]
       }
     }
   }
   

3. Save and restart Cline.
4. Verify that the MCP server is running by listing active MCP servers in Cline.

Securing API Keys

• Use environment variables to store sensitive data, such as API keys.
• Example configuration:

  {
    "mcpServers": {
      "browser-use": {
        "command": "npx",
        "args": ["@browser-use/mcp-server@latest"],
        "env": {
          "API_KEY": "${API_KEY}"
        },
        "inputs": {
          "api_key": "${API_KEY}"
        }
      }
    }
  }
  

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "browser-use": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change “browser-use” to whatever the actual name of your MCP server is and replace the URL with your own MCP server URL.

Overview

Between the two tables:
This MCP server provides the essentials for browser automation in an AI context and is well-maintained, but lacks in-depth documentation on available prompts, resources, and tools. For core usage (browser control), it is highly valuable, but documentation completeness holds it back.

MCP Score

Overall rating:
6/10 (excellent for core browser automation, but documentation on advanced MCP concepts and tooling is lacking; would be higher with more implementation details exposed).