Home / MCP / tailscale MCP Server

tailscale MCP Server

Provides an MCP interface to query Tailscale status by running the tailscale CLI on macOS.

python

6stars

View on GitHub

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
    "mcpServers": {
        "tailscale": {
            "command": "python",
            "args": [
                "tailscale.py"
            ]
        }
    }
}

This MCP server lets you query Tailscale status by running the tailscale CLI on macOS, exposing a lightweight interface you can connect to from your MCP clients. It’s useful for integrating real-time Tailscale state into your tooling and automation workflows without duplicating CLI logic in your client.

How to use

You start the MCP server locally and then connect your MCP client to it to fetch Tailscale status. By default, the server runs in STDIO mode, which means it communicates via the standard input and output streams of the process you started. If you choose the HTTP transport option, you can also connect over HTTP to a local port. The server relies on the tailscale CLI being installed and available at the expected macOS path.

How to install

Prerequisites you need to satisfy before running the server:

Python is installed on your system
Tailscale is installed and accessible at /Applications/Tailscale.app/Contents/MacOS/Tailscale
uv is available for dependency management

Step by step flow you can follow to get the server running locally:

# Step 1: Prepare the working directory
# Create a directory for the MCP server and place tailscale.py inside it

# Step 2: Start the server using the default STDIO transport
python tailscale.py

# Optional: start the server with HTTP transport on a specific port
# This enables HTTP-based MCP client connections
python tailscale.py --transport http --port 4001