# System Diagnostics MCP Server

This is a Python-based Model Context Protocol (MCP) server. It provides diagnostic capabilities:
1. `get_system_metrics`: Retrieves real-time CPU, Memory, and Disk usage metrics.
2. `list_top_processes`: Lists top running processes sorted by CPU or Memory usage.
3. `check_port_status`: Verifies if a specific TCP port is open on a given host.

## Prerequisites

- Python 3.10+
- [uv](https://github.com/astral-sh/uv) (for ultra-fast dependency management)
- Docker (optional, for containerized execution)

## Local Setup (Using uv)

1. Clone the repository and navigate into the folder:
   ```bash
   git clone https://github.com/niketanjain/mcp-server-setup.git
   cd mcp-server-setup
   ```

2. Initialize the virtual environment and install dependencies:
   ```bash
   uv sync
   # OR if no lockfile:
   uv pip install mcp psutil pydantic starlette uvicorn
   ```

3. Run the server locally:
   ```bash
   uv run python server.py
   ```
   The server will start an HTTP/SSE server on `http://localhost:6500/sse`.

## Running with Docker

1. Build the Docker image:
   ```bash
   docker build -t mcp-sysdiag-server .
   ```

2. Run the Docker container, exposing port 6500:
   ```bash
   docker run -p 6500:6500 mcp-sysdiag-server
   ```

## Extending to Production

Check out [production-grade.md](./production-grade.md) for recommendations and best practices when extending this MCP server to a production environment.