tesla-fleet-mcp
Minimal stdio MCP server for Tesla Fleet API, enabling vehicle listing, state checks, live data, nearby chargers, wake, and command tools (climate, charge, lock) via optional tesla-http-proxy.
README
tesla-fleet-mcp
Minimal stdio MCP for Tesla Fleet API. TypeScript. One process, env-based secrets, no extra daemons for read/wake.
Commands (climate, charge, lock) need Tesla’s official tesla-http-proxy plus a virtual key on the car. That proxy is not bundled here.
Hermes server name: tesla-fleet. Tools appear as mcp__tesla_fleet__*.
Tools
| Tool | Notes |
|---|---|
vehicles_list |
Account vehicles |
vehicle_get |
Cheap state (online / asleep / offline) |
vehicle_data |
Live data. Billable. Default = full payload (no endpoints filter). Optional endpoints list; Tesla wants ; (commas are accepted by this MCP and rewritten). location_data for GPS |
nearby_chargers |
Nearby charge sites |
wake_up |
Wake from sleep |
climate_start / climate_stop / set_temps |
Needs command proxy |
charge_start / charge_stop / set_charge_limit |
Needs command proxy |
door_lock / door_unlock |
Needs command proxy |
Do not poll vehicle_data. Tesla bills per call. Check vehicle_get first.
A comma-separated endpoints query returns metadata only (no charge_state). This server rewrites commas to ;.
Requirements
- Node 20+
- A Tesla developer application (dashboard)
- An HTTPS hostname you own (must not contain the word
tesla) - Public key at
https://<your-domain>/.well-known/appspecific/com.tesla.3p.public-key.pem - Partner
POST /api/1/partner_accountsin each region you use - Payment method + billing limit (Fleet API is pay-as-you-go; ~$10/month credit)
Environment
Copy env.example and point TESLA_CACHE_PATH at a 0600 file outside git.
| Variable | Required | Meaning |
|---|---|---|
TESLA_CLIENT_ID |
yes | Developer app client id |
TESLA_CLIENT_SECRET |
yes | Developer app secret |
TESLA_REDIRECT_URI |
yes | Must match the app’s allowed redirect (e.g. https://your.domain/callback) |
TESLA_AUDIENCE |
yes | Fleet base URL for your region (EU example in env.example) |
TESLA_CACHE_PATH |
no | Token cache path (default: ./token-cache.json) |
TESLA_REGION |
no | eu (default) or na |
TESLA_VIN |
no | Default vehicle if tools omit vin |
TESLA_COMMAND_BASE |
no | Proxy origin, e.g. https://127.0.0.1:4443 |
NODE_EXTRA_CA_CERTS |
no | Proxy TLS CA if you use a self-signed localhost cert |
Never commit secrets. Never paste refresh tokens or client secrets into chat.
Setup (once)
- Create the Tesla developer app. Grant type: authorization code and machine-to-machine.
- Allowed origin:
https://your.domain. Allowed redirect:https://your.domain/callback. - Scopes you need, typically:
openid,offline_access,vehicle_device_data, plusvehicle_cmds/vehicle_charging_cmds/vehicle_locationif you use those tools. - Generate a P-256 key pair:
openssl ecparam -name prime256v1 -genkey -noout -out private-key.pem
openssl ec -in private-key.pem -pubout -out public-key.pem
Host only the public key at:
https://your.domain/.well-known/appspecific/com.tesla.3p.public-key.pem
- Register the partner account (EU example):
# partner token
curl -s --data-urlencode grant_type=client_credentials \
--data-urlencode client_id="$TESLA_CLIENT_ID" \
--data-urlencode client_secret="$TESLA_CLIENT_SECRET" \
--data-urlencode audience="$TESLA_AUDIENCE" \
--data-urlencode scope='openid vehicle_device_data vehicle_cmds' \
https://fleet-auth.prd.vn.cloud.tesla.com/oauth2/v3/token
# then POST {"domain":"your.domain"} to $TESLA_AUDIENCE/api/1/partner_accounts
# with Authorization: Bearer <partner access_token>
- User login (owner of the car):
set -a && source .env && set +a
npm install
npm run login
# open the printed URL, approve, paste the full https://your.domain/callback?code=... URL:
npm run login -- "https://your.domain/callback?code=...."
The callback host only needs to accept the browser hit; this CLI reads the URL you paste. A blank page is fine.
Hermes
# example wrapper: source your env file, then tsx
hermes mcp add tesla-fleet --command /path/to/tesla-fleet-mcp/run.sh
run.sh sources .env in the repo directory or $TESLA_ENV if set. Restart the Hermes process/gateway after adding. Enable/disable tools in the client (tools.include / exclude), not with extra server flags.
Commands (optional)
Model 3 / Y and recent S/X require the Vehicle Command Protocol. Unsigned Fleet command calls are rejected.
- Run
tesla-http-proxywith your private key, TLS, localhost only. - Pair the virtual key (Tesla app, car online):
https://tesla.com/_ak/your.domain - Set
TESLA_COMMAND_BASE=https://127.0.0.1:4443(andNODE_EXTRA_CA_CERTSif the proxy cert is self-signed).
Do not publish the proxy to the internet.
Dev
npm test
npx tsc --noEmit
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.