MCP Client

What is MCP Client?

An interactive AI assistant interface with a dual-display system: a terminal-based CLI for commands and an always-on-top window for real-time model output monitoring. It runs on desktop environments with Node.js and Python, and connects to MCP servers for enhanced AI capabilities.

How to use MCP Client?

Install Node.js, npm, and Python 3.7+. Clone the repository, run npm install and pip install backend/requirements.txt and pip install mcp. Start the application with python backend/cli.py. Configure an MCP server in backend/server.config to enable AI integration.

Key features of MCP Client

• Terminal-based CLI for AI assistant interaction
• Always-on-top translucent display window
• Real-time model output monitoring
• Pause functionality for model execution
• Global shortcuts: toggle follow desktop and pause/resume
• MCP server integration for enhanced capabilities

Use cases of MCP Client

• Control your computer with AI assistance
• Navigate and interact with web browsers while monitoring AI actions
• Monitor progress and provide input during long-running tasks
• Observe and control AI actions in real-time for safety

FAQ from MCP Client

What are the system requirements?

Node.js (v12 or higher), npm, and Python 3.7+.

How does MCP server integration work?

Configure your MCP server in backend/server.config, install the mcp Python library, and start the application. The app automatically connects to the specified server.

What license is MCP Client under?

The application is licensed under ISC.

What happens if the MCP server is unavailable?

The application falls back to a mock LLM implementation if the MCP server is not available.