Browser MCP provides an AI-driven browser automation solution, enabling applications like VS Code, Claude, and Cursor to control a user's existing browser. This offers a private, fast, and stealthy approach to web automation by leveraging local processing and real browser fingerprints, benefiting AI developers and power users seeking seamless integration with their online workflows.

How It Works

Browser MCP functions as a Model Context Provider (MCP) server coupled with a Chrome extension. It adapts the Playwright MCP server architecture to directly control the user's active browser instance, rather than launching new ones. This design choice allows it to utilize existing browser profiles, preserving logged-in sessions and bypassing common bot detection mechanisms by mimicking a real browser fingerprint.

Quick Start & Requirements

• Installation requires building from the monorepo due to unreleased dependencies.
• Requires a Chrome browser.
• Further setup details are available in the Docs.

Highlighted Details

• Local automation for enhanced performance and privacy.
• Utilizes existing browser profiles for logged-in sessions.
• Employs real browser fingerprints to avoid bot detection and CAPTCHAs.
• Adapted from the Playwright MCP server for direct browser control.

Maintenance & Community

• Development is ongoing within a monorepo structure.
• Community and contribution details are not explicitly detailed in the README.

Licensing & Compatibility

• The license is not specified in the README.
• Compatibility for commercial use or closed-source linking is not detailed.

Limitations & Caveats

The project is currently not independently buildable due to dependencies on unreleased code within its monorepo. Specific licensing terms and commercial use compatibility are not yet defined.