AI Assistant MCP

What is AI Assistant MCP?

AI Assistant MCP is a powerful AI assistant with system monitoring and control capabilities, built with FastAPI and Ollama. It provides an interactive chat interface alongside real-time system metrics, file operations, and process management, all accessible via a modern web interface and RESTful API.

How to use AI Assistant MCP?

Set up by cloning the repository, creating a Python virtual environment, installing dependencies, and adding a SECRET_KEY to your .env file. Start the Ollama service with ollama serve, then run python run.py. Access the web interface at http://localhost:8000 to chat with the AI, monitor system resources, manage files and processes, and view alerts. Interact programmatically through the documented API endpoints.

Key features of AI Assistant MCP

• Interactive AI chat with context-aware conversations and history management.
• Real-time system monitoring dashboard (CPU, memory, disk, network).
• Process tracking, management, and resource usage alerts.
• File system operations (list, read, write) and process control.
• Secure API key authentication for all endpoints.
• Modern, responsive web UI with real-time updates.

Use cases of AI Assistant MCP

• Query system resource usage (CPU, memory, disk, network) via natural language.
• Manage files and processes remotely through chat commands or API calls.
• Automate system health checks and receive alerts on resource thresholds.
• Build custom tools that combine AI assistance with direct system control.

FAQ from AI Assistant MCP

How is authentication handled?

All API endpoints require API key authentication. Configure the key in your .env file under SECRET_KEY.

What are the system requirements?

You need Python, Ollama (to run the AI model), and the dependencies listed in requirements.txt. Ollama must be running locally before starting the application.

How do I start the AI Assistant MCP server?

Clone the repo, install dependencies, set environment variables, start Ollama with ollama serve, and then run python run.py. The web interface is available at http://localhost:8000.

What API endpoints are available?

Chat: POST /chat. Monitoring: GET /monitoring, /monitoring/metrics, /monitoring/history, /monitoring/alerts. System control: endpoints for listing/reading/writing files and listing/getting/terminating processes.

Can I monitor system resources in real time?

Yes. The monitoring dashboard provides real-time metrics (CPU, memory, disk, network) with interactive charts and visualizations, plus resource usage alerts.