home / mcp / 302 browser use mcp server

302 Browser Use MCP Server

302 BrowserUse MCP

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "302ai-302_browser_use_mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@302ai/browser-use-mcp"
      ],
      "env": {
        "302AI_API_KEY": "YOUR_API_KEY_HERE"
      }
    }
  }
}

You can run an MCP server that lets you control a browser automation tool through natural language, enabling remote tasks and research workflows. This server exposes an easy way to start browser tasks, check their status, and integrate with MCP-enabled clients for seamless control across tools and platforms.

How to use

You interact with the 302AI Browser Use MCP Server through an MCP client. Set your client to connect using the stdio method by launching the server with the provided command, then issue high-level actions such as creating a browser automation task or querying task status. You can integrate with popular MCP clients like Claude Desktop, Cherry Studio, or ChatWise by configuring an MCP server entry that points to the 302ai package.

Typical usage patterns include starting a browser automation task from a natural language request and then polling for its status or results. The server is designed to dynamically load its tool list from a remote source, so you can access new capabilities without manual updates. Use the standard MCP workflow your client supports to submit tasks, monitor progress, and retrieve results.

How to install

Prerequisites: you need Node.js and npm installed on your system.

Install dependencies for the server package.

Build the server to prepare it for runtime.

Optionally start a development watcher for auto-rebuilds during development.

# Install dependencies
npm install

# Build the server
npm run build

# Development with auto-rebuild (optional)
npm run watch

Additional notes

Configure your MCP client with the following stdio entry to run the server locally via npx and pass your API key needed by the 302AI service.

{
  "mcpServers": {
    "browser_use_mcp": {
      "command": "npx",
      "args": ["-y", "@302ai/browser-use-mcp"],
      "env": {
        "302AI_API_KEY": "YOUR_API_KEY_HERE"
      }
    }
  }
}

Troubleshooting and debugging

Debugging MCP servers can be challenging because communication occurs over stdio. Use the MCP Inspector to simplify debugging. Run the inspector command to start a debugging session and obtain a URL for browser-based tools.

npm run inspector

Security and maintenance

Protect your API key and manage access to the MCP server configuration. Use a dedicated API key with scoped permissions and rotate keys periodically. Monitor the remote tool list as it can update dynamically, and verify that new tools meet your security and compliance requirements.

Available tools

Create Browser Automation Task

Submit a request to start a new browser automation task based on natural language input. The server initializes a browser automation workflow and returns a task identifier and initial status.

Query Browser Task Status

Check the current status and progress of an existing browser automation task by its identifier, returning updates and results as they become available.