jetson-mcp

Name: jetson-mcp
Author: Zalmotek

@Zalmotek

Repository

About jetson-mcp

A MCP server for using natural language to monitor and remote control a Nvidia Jetson board

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "jetson-mcp": {
      "command": "python",
      "args": [
        "app/main.py"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is jetson-mcp?

jetson-mcp is an MCP (Model Context Protocol) server that lets you use natural language to monitor and remotely control a Nvidia Jetson board from clients on the same network. It exposes tools over SSE (Server-Sent Events) transport, reading hardware info from /etc/nv_boot_control.conf and software info from /etc/nv_tegra_release and /proc/version. It is built with the FastMCP library.

How to use jetson-mcp?

Clone the repo on the Jetson, run ./install.sh to set up a Python virtual environment, then run sudo ./setup_service.sh to create a systemd service listening on port 8000. Start the service with sudo systemctl start jetson-mcp.service. From a remote MCP client, connect to the SSE endpoint at http://<jetson_ip>:8000/sse.

Key features of jetson-mcp

Provides MCP tools via SSE transport on the network.
get_jetson_hw_info: reads module/carrier board info from config file.
get_jetson_sw_info: reads Jetpack version and Linux kernel version.
Includes scripts for easy installation and systemd service setup.
Requires firewall to allow incoming TCP on port 8000.

Use cases of jetson-mcp

Remotely query Jetson hardware configuration from a desktop AI client.
Check Jetpack and kernel version on a Jetson without SSH or terminal access.
Integrate Jetson monitoring into an LLM-powered automation workflow.
Prototype edge AI applications with natural language control over the board.

FAQ from jetson-mcp

How does jetson-mcp expose its tools?

It uses SSE (Server-Sent Events) transport on port 8000, not the stdio transport typical of local MCP servers. Clients connect via http://<jetson_ip>:8000/sse.

What are the runtime dependencies?

The server runs in a Python virtual environment created by install.sh. It requires the FastMCP library (installed from requirements.txt) and reads system files that exist on a standard Nvidia Jetson Linux installation.

Can I use jetson-mcp with Cursor?

Cursor’s mcp.json is designed for local stdio transport, so connecting to this networked SSE server may require a proxy or different configuration steps. Consult your client’s documentation for SSE endpoint support.

How do I start jetson-mcp automatically on boot?

The setup_service.sh script creates and enables a systemd service file (/etc/systemd/system/jetson-mcp.service). After running sudo systemctl enable jetson-mcp.service, the server starts automatically on boot.