Tesla

You can run an MCP server that provides access to your Tesla Fleet data and actions through any MCP-capable assistant or agent. This server lets you list your vehicles, wake them, get real-time location data with a Google Maps link, and refresh or debug your vehicle list, all while keeping credentials securely in your control.

How to use

You will connect your MCP client to the Tesla MCP Server using either the hosted HTTP/SSE endpoint or a local stdio setup. After connecting, you can perform operations such as listing your vehicles to obtain IDs, waking a vehicle from sleep, and querying the current location. Each tool can be called with the appropriate vehicle identifier and will return structured data that you can display or act upon in your assistant.

How to install

Prerequisites: ensure you have Node.js installed on your machine. You may also choose the hosted option which requires no local installation.

Option 1 — Hosted HTTP/SSE server (recommended for quick setup) follow these steps to use the hosted service.

1. Add the server in your MCP client using the following URL.

2. Complete the first-time connection flow by obtaining a setup URL, entering your Tesla Developer credentials, and then using the connection URL from the success page as your MCP server URL to reconnect.

3. If you need to register your Tesla credentials, you’ll set the redirect URI in your Tesla developer app to your server’s callback. Use the provided redirect URI for your environment.

Option 2 — Local HTTP server (multi-user, browser-based authentication). Install and run locally.

# Option 1: Hosted HTTP/SSE server
# No local installation required; connect to the hosted endpoint
# Server URL to use in your MCP client:
https://tesla-mcp.onrender.com/sse

Option A: HTTP server (multi-user, browser auth)

Clone the project, install dependencies, and start the HTTP server.

git clone https://github.com/Sara3/Tesla-MCP.git
cd Tesla-MCP
npm install
npm run build
npm run start:http

Option B: Stdio (single user, .env only)

For a single user with credentials stored in a .env file, provide the Tesla credentials and run the server in stdio mode.

# .env
TESLA_CLIENT_ID=...
TESLA_CLIENT_SECRET=...
TESLA_REFRESH_TOKEN=...

npm run build
npm start

Configure your MCP client to run the server command, for example: node run-mcp.js. Obtain a refresh token with npm run get-token.

## Available tools

### get\_setup\_url

Obtain the URL to set up Tesla Developer credentials so users can authorize access.

### get\_auth\_url

Get the URL to connect the Tesla account after initial setup.

### list\_vehicles

List your vehicles and their IDs to be used with other MCP tools.

### get\_vehicle\_location

Return the current location of a vehicle, including lat/long and a Google Maps link.

### wake\_up

Wake a vehicle from sleep so you can query data or control it.

### refresh\_vehicles

Refresh the list of vehicles from the API to reflect any changes.

### debug\_vehicles

Provide debug information such as IDs, VINs, and vehicle state for troubleshooting.