TuriX CUA
This is the official website for TuriX Computer-use-Agent
What is TuriX CUA?
TuriX CUA is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to this is the official website for turix computer-use-agent
This is the official website for TuriX Computer-use-Agent
This server falls under the Browser Automation category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- This is the official website for TuriX Computer-use-Agent
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx turix-cuaConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use TuriX CUA
TuriX CUA is an open-source, state-of-the-art Computer-Use Agent that lets AI models take real, hands-on actions directly on your desktop — clicking, typing, and navigating GUI applications just as a human would. It achieves an 80%+ success rate on macOS desktop automation benchmarks and ranks 3rd on OSWorld with 64.2%, making it one of the most capable open desktop agents available. Developers and researchers use it to automate complex multi-step workflows across macOS, Windows, and Linux without needing app-specific APIs, and it connects to Claude and other MCP-compatible clients via the Model Context Protocol.
Prerequisites
- macOS 15+ (primary), Windows, or Linux
- Python 3.12 environment (conda recommended)
- An MCP client such as Claude Desktop
- macOS Accessibility and Safari Automation permissions granted to Terminal/IDE
- An AI model API key (TuriX API, Qwen3-VL, or other supported VLM configured in config.json)
Clone the repository
Clone the TuriX CUA repository from GitHub. For Windows use the multi-agent-windows branch; for Linux use multi-agent-linux.
git clone https://github.com/TurixAI/TuriX-CUA.git
cd TuriX-CUACreate and activate a Python 3.12 environment
TuriX requires Python 3.12. Use conda to create an isolated environment and install dependencies.
conda create -n turix_env python=3.12
conda activate turix_env
pip install -r requirements.txtGrant macOS permissions
Open System Settings > Privacy & Security > Accessibility and add Terminal, your IDE, and /usr/bin/python3. Also enable Safari Automation under Privacy & Security > Automation if you want web tasks.
Configure the model in config.json
Edit config.json in the project root to set your chosen VLM provider and API key. TuriX supports its own API, Qwen3-VL, and other hot-swappable models without touching any other code.
Add TuriX as an MCP server in Claude Desktop
Register TuriX as an MCP server so Claude Desktop can invoke desktop automation tools. Use npx turix-cua as the command in your MCP config.
Run TuriX and verify operation
Start TuriX manually to confirm the agent can control the desktop, then switch to launching it via your MCP client.
python -m turixTuriX CUA Examples
Client configuration
Add TuriX CUA to Claude Desktop's MCP server list so Claude can issue desktop automation tasks.
{
"mcpServers": {
"turix-cua": {
"command": "npx",
"args": ["turix-cua"]
}
}
}Prompts to try
Example natural-language instructions you can send through Claude once TuriX CUA is connected.
- "Search for the latest AI news in Safari and create a Pages document summarising the top 5 stories, then send it to John via Messages."
- "Open the Numbers file my boss sent me, generate a bar chart from the sales data, and insert it into the PowerPoint on my Desktop."
- "Book the cheapest non-stop flight from SFO to JFK next Tuesday using the TuriX browser."Troubleshooting TuriX CUA
Agent fails to click or control GUI elements
Make sure Terminal (or your IDE) is listed under System Settings > Privacy & Security > Accessibility. If you run via VS Code, add VS Code explicitly. Also add /usr/bin/python3 to the Accessibility list.
Model not responding or returning empty actions
Check config.json to ensure the correct API key and model endpoint are set. TuriX ships with its own API model by default; if you switch to Qwen3-VL or another VLM, verify the model name string matches the provider's documentation.
npx turix-cua command not found in MCP client
Ensure Node.js 18+ is installed and that the turix-cua package is reachable via npx. Alternatively, run TuriX directly with Python (python -m turix) and adjust your MCP config to use the python command with the correct conda environment path.
Frequently Asked Questions about TuriX CUA
What is TuriX CUA?
TuriX CUA is a Model Context Protocol (MCP) server that this is the official website for turix computer-use-agent It connects AI assistants to external tools and data sources through a standardized interface.
How do I install TuriX CUA?
Follow the installation instructions on the TuriX CUA GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with TuriX CUA?
TuriX CUA works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is TuriX CUA free to use?
Yes, TuriX CUA is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
TuriX CUA Alternatives — Similar Browser Automation Servers
Looking for alternatives to TuriX CUA? Here are other popular browser automation servers you can use with Claude, Cursor, and VS Code.
Chrome DevTools MCP
★ 40.6kAI-powered Chrome automation server with natural language element detection. Control Chrome browser through MCP protocol for testing, debugging, and performance analysis. Features 91% accuracy in element location, works with free AI models, and suppo
UI TARS Desktop
★ 34.9k📇 🏠 - Browser automation capabilities using Puppeteer, both support local and remote browser connection.
Playwright
★ 32.8kA production-ready browser automation server that enables AI assistants to interact with web pages using tools for navigation, element interaction, and data extraction. It features a built-in Inspector UI and robust crash recovery for reliable automa
Page Agent
★ 18.0kJavaScript in-page GUI agent. Control web interfaces with natural language.
Chrome
★ 11.7kAn extension-based MCP server that enables AI assistants to control your browser, leveraging existing sessions and login states for automation and content analysis. It provides over 20 tools for semantic tab search, interactive element manipulation,
LAMDA
★ 7.8kThe most powerful Android RPA agent framework, next generation mobile automation.
Browse More Browser Automation MCP Servers
Explore all browser automation servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up TuriX CUA in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use TuriX CUA?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.