Browser MCP

Browser MCP is a Model Context Protocol (MCP) server that connects LLM agents to Chromium instances. It provides tools for web page interaction, cookie retrieval, form inputs, and screenshot capture, allowing AI assistants to execute browser tasks.

Key Features of Browser MCP

• Direct DOM Manipulation: Exposes tools for clicking elements, entering text, and selecting options via MCP.
• State Retention: Maintained browser sessions allow agents to execute complex, multi-page flows.
• Visual Snapshot Captures: Returns PNG screenshots of page layouts directly to the calling AI model.
• Auth Credentials Sharing: Seamlessly re-uses local browser sessions to bypass verification barriers during tasks.

Benefits of Using Browser MCP

• Autonomous Web Tasks: Equips AI assistants to purchase products, fill forms, and gather web data automatically.
• Frictionless Scripting: Automate browser workflows using conversational prompts rather than writing complex scripts.
• Secure Local Run: Operates within your local security profile, ensuring you maintain complete control over actions.

To allow AI coding agents to assist in QA, Browser MCP connects local browser sessions directly to LLMs, letting them debug issues in real-time.