simple-mcp-selenium

brutalzinn/simple-mcp-selenium

3.1

If you are the rightful owner of simple-mcp-selenium and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

MCP Selenium Server is a powerful Model Context Protocol (MCP) server designed for comprehensive browser automation using Selenium WebDriver.

Tools
5
Resources
0
Prompts
0

MCP Selenium Server

Powerful browser automation for Cursor IDE with multi-browser instance support

Documentation GitHub

🚀 What is this?

MCP Selenium Server lets you control browsers using natural language in Cursor IDE. Perfect for testing, automation, and multi-user scenarios.

Example: Just tell Cursor "Open a browser, go to Google, search for 'hello world', and take a screenshot" - and it happens!

⚡ Quick Start

1. Install (Docker - Recommended)

git clone https://github.com/brutalzinn/simple-mcp-selenium.git
cd simple-mcp-selenium
./scripts/install-simple.sh

2. Configure Cursor IDE

Add to Cursor settings:

{
  "mcp.servers": {
    "selenium-browser": {
      "command": "docker",
      "args": [
        "exec",
        "-i",
        "mcp-selenium-server",
        "node",
        "/app/dist/index.js"
      ]
    }
  }
}

3. Use it!

Restart Cursor and start talking to your browser:

  • "Open a browser and go to example.com"
  • "Click the login button"
  • "Type 'admin@example.com' in the email field"
  • "Take a screenshot"

🎯 Key Features

  • 🤖 Natural Language Control - Tell Cursor what to do in plain English
  • 🌐 Multi-Browser Support - Manage multiple browser instances with unique IDs
  • 🔌 Plugin System - Extend functionality with custom plugins
  • 📸 Screenshots - Capture page states automatically
  • 🔄 Drag & Drop - Full interaction support
  • ⚡ Fast - Optimized for speed (3-second timeouts)

📚 Documentation

Complete documentation: https://brutalzinn.github.io/simple-mcp-selenium/

🛠️ Available Tools

ToolWhat it doesExample
open_browserStart browser"Open a browser"
navigate_toGo to URL"Go to google.com"
click_elementClick things"Click the login button"
type_textType text"Type 'hello' in the search box"
drag_and_dropMove elements"Drag the menu to the workflow area"
take_screenshotCapture page"Take a screenshot"
execute_scriptRun JavaScript"Execute some JavaScript"

🌐 Multi-Browser Mode

Perfect for testing multiple users or scenarios:

# User 1
"Open a browser with ID 'user1' and go to site1.com"

# User 2
"Open a browser with ID 'user2' and go to site2.com"

# Use specific browsers
"Click the button in browser 'user1'"
"Fill the form in browser 'user2'"

🔌 Plugins

Extend functionality with plugins:

View all plugins →

🐳 Docker Commands

# Start
docker-compose up -d

# Stop
docker-compose down

# Restart
docker-compose restart

# View logs
docker-compose logs -f

🤝 Contributing

  1. Fork the repo
  2. Create your feature branch
  3. Make your changes
  4. Submit a pull request

📄 License

MIT License - see for details.

🆘 Need Help?


Made with ❤️ for the Cursor IDE community

Made with 😠 for other alternatives that uses internal proxies to somewhere i dont know