Agent-Browser MCP

English | 简体中文 | 日本語

Model Context Protocol (MCP) server for agent-browser - providing complete browser automation capabilities for AI agents.

This project is an independent MCP server implementation that wraps the excellent agent-browser CLI tool, making its powerful browser automation features available through the Model Context Protocol.

Features

• 🔧 44 Tools - Complete coverage of agent-browser's functionality
• 🎯 Token-Efficient @ref System - Reduces token usage by caching element references
• 🌐 Full Playwright API - Leverage the complete browser automation capabilities
• 🔄 Auto-Launch - Browser starts automatically when needed
• 💾 State Persistence - Save and restore browser state across sessions
• 🎬 Video Recording - Record browser sessions for debugging
• 🌐 Network Interception - Monitor and modify network requests
• 📊 Session Management - Manage multiple tabs and windows

Installation

Using npm

npm install agent-browser-mcp-server

From Source

git clone https://github.com/hughedward/agent_browser_mcp.git
cd agent_browser_mcp
npm install
npm run build

Quick Start

For Claude Desktop

1. Install the package
2. Configure in Claude Desktop settings (~/.claude/settings.json):

{
  "mcpServers": {
    "agent-browser-mcp-server": {
      "command": "npx",
      "args": ["agent-browser-mcp-server"],
      "env": {
        "HEADED": "false"
      }
    }
  }
}

Standalone

agent-browser-mcp-server

Available Tools

Core Tools

• browser_navigate - Navigate to a URL
• browser_snapshot - Capture page structure with @ref system
• browser_screenshot - Take screenshots
• browser_close - Close browser/page

Navigation & History

• browser_back - Go back in history
• browser_forward - Go forward in history
• browser_reload - Reload the current page

Element Interaction

• browser_click - Click an element
• browser_fill - Fill input fields
• browser_type - Type without clearing
• browser_select - Select dropdown options
• browser_check / browser_uncheck - Check/uncheck checkboxes
• browser_drag - Drag and drop
• browser_upload - Upload files
• browser_dblclick - Double click
• browser_focus - Focus elements
• browser_hover - Hover over elements
• browser_scroll - Scroll page
• browser_press - Press keyboard keys

Element Discovery

• browser_find - Semantic element search (role, text, label, placeholder, etc.)
• browser_get - Get element information
• browser_is - Check element state

Tabs & Windows

• browser_tab - Manage tabs
• browser_window - Manage windows
• browser_frame - Switch to iframes

Advanced Features

• browser_record - Record browser sessions
• browser_network - Monitor network requests
• browser_console - Access console
• browser_errors - Track JavaScript errors
• browser_trace - Performance tracing
• browser_profiler - Chrome DevTools profiling
• browser_evaluate - Execute JavaScript
• browser_pdf - Export to PDF
• browser_dialog - Handle JavaScript dialogs
• browser_download - Manage downloads

State & Storage

• browser_state - Save/load browser state
• browser_cookies - Manage cookies
• browser_storage - Access localStorage/sessionStorage

Utilities

• browser_wait - Wait for conditions
• browser_set - Set element attributes
• browser_mouse - Mouse control
• browser_diff - Compare pages
• browser_highlight - Debug highlighting

Configuration

Environment Variables:

Development

# Install dependencies
npm install

# Build
npm run build

# Run in development mode (auto-rebuild)
npm run dev

# Run tests
npm test

# Watch mode
npm run test:watch

# Start server
npm start

Documentation

• CLAUDE.md - Development guide for Claude Code
• TESTING_GUIDE.md - Testing instructions
• QUICK_TEST_GUIDE.md - Quick reference

Related Projects

• agent-browser - Original CLI tool this project wraps
• Model Context Protocol - The protocol this server implements

License

Apache-2.0

Note: This project is an independent implementation and is not officially affiliated with Vercel or the original agent-browser project.