Browser Control MCP is a free server tool by Eyalzh, built with Node.js and TypeScript, designed to enable AI agents and assistants to interact with the Firefox browser. It functions as an MCP server for browser automation, connecting to the user's Firefox instance via a dedicated companion extension. This tool is distributed as source code, a Docker image, a Firefox extension, and a Claude Desktop DXT file, making it accessible for various development environments. It provides a programmatic interface for managing browser activities.

What does Browser Control MCP connect to?

Browser Control MCP connects directly to a user's Firefox browser. This connection is established and maintained through a companion browser extension, which must be installed within Firefox. The system operates as an MCP server, specifically categorized for browser automation and control. This architecture allows AI clients to send commands to the server, which then relays them to the Firefox browser via the installed extension, enabling remote interaction and management of browser functions.

Core features and tools

Browser Control MCP exposes several callable tools for browser interaction. These include the ability to Open or close tabs, Get the list of opened tabs, and Reorder opened tabs. Users can also Create tab groups with name and color. For information retrieval, it can Read and search the browser's history, and Read a webpage's text content and links (requiring user consent). Advanced visual tools allow it to Find and highlight text in a browser tab (with user consent) and Capture a screenshot of a tab's visible area (requiring user consent per tab).

Who is it for?

This program is intended for AI agents and AI assistants, particularly developers who utilize AI coding assistants such as Augment Code or Claude Desktop. It provides the necessary tools for these AI entities to perform browser-based tasks, from managing tabs to conducting research. The design supports automated workflows that require direct interaction with a web browser. Skip this if you do not work with AI agents or AI assistants.

How to get started

The server authenticates using an API key passed as an environment variable. This token must also be configured in the companion browser extension's options page, and the server uses HTTP as a transport protocol on a configurable port like PORT=3000.

Use cases

• Tab management: For example, "Close all non-work-related tabs in my browser." or "Group all development related tabs in my browser into a new group called 'Development'."
• Browser history search: Such as "Help me find an article in my browser history about the Milford track in NZ."
• Browsing and research: For instance, "In my browser, use Google Scholar to search for papers about L-theanine in the last 3 years. Open the 3 most cited papers. Read them and summarize them for me."

Browser Control MCP offers a specialized set of tools for AI agents to interact with Firefox. Its architecture, relying on a companion extension, enables precise browser automation for tasks like tab management, history search, and web content analysis. The program supports HTTP transport and integrates with AI clients like Claude Desktop. It is a capable solution for developers building AI-driven browser workflows. Try using its tab grouping feature to organize your browser workspace.