Enables generating custom Tesla vehicle wrap designs using Tesla's official UV templates and panel stencils, with tools for listing supported vehicles, fetching templates, creating wrap images, verifying them against the vehicle mask, and flattening outputs.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@tesla-wrap Generate a black and gold AI wrap design for my Model 3 Highland."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Tesla Paint Shop wrap generator: official UV template + Grok Imagine + panel stencil.
Works for every vehicle Tesla publishes on teslamotors/custom-wraps.
Client usage (what an MCP host should follow): USAGE.md, also wrap://usage.
Download the official UV for the vehicle
(Optional) seed from a geometric layout PNG
Grok Imagine paints the look at 2k / medium quality
Stencil through Tesla’s panel mask → 1024×1024 PNG under 1 MB
Imagine is for look. Outlines always come from Tesla’s template (IoU 1.0).
Related MCP server: SudoMock
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
cp .env.example .envEdit .env:
XAI_API_KEY — required for Imagine
MCP_LOGIN_PASSWORD — required if you run the public HTTP MCP
.env is gitignored. Do not put keys or passwords in this repo.
Official templates download on first fetch / create_wrap. Generated PNGs land in outputs/ (also gitignored).
python wrap.py vehicles
python wrap.py fetch highland
python wrap.py imagine --name black-gold --vehicle highland \
--look "driver side chrome gold, passenger matte black, honeycomb fade on the hood" \
--ref https://example.com/car.jpg
python wrap.py verify outputs/black-gold.png --vehicle highland
python wrap.py flatten outputs/black-gold.png--base path/to/layout.png restyles an already-laid-out UV instead of the blank template.
Tools: list_vehicles, fetch_template, create_wrap, verify_wrap, flatten_wrap.
[mcp_servers.tesla-wrap]
command = "/path/to/grok-editor/.venv/bin/python"
args = ["/path/to/grok-editor/mcp_server.py"]python mcp_server.py --http --host 127.0.0.1 --port 8787 --path /grok-editor-mcpSet MCP_PUBLIC_HOST to the hostname you funnel/proxy (no https://). First connect opens /login; the password is MCP_LOGIN_PASSWORD from .env. After that the client keeps an OAuth token. Imagine still uses local XAI_API_KEY.
[mcp_servers.tesla-wrap]
url = "https://YOUR_HOST/grok-editor-mcp"
enabled = true
tool_timeout_sec = 600Health (no auth): http://127.0.0.1:8787/health
Optional systemd user unit: tesla-wrap-mcp.service (WorkingDirectory uses %h/grok-editor).
model3, model3-2024-base (aliases: highland, model3-premium), model3-2024-performance, modely, modely-2025-base, modely-2025-premium, modely-2025-performance, modely-l, models-2021, models-2025-plaid, modelx-2021, cybertruck.
This server cannot be deployed
Auto.dev MCP — wraps the Auto.dev automotive data API (auto.dev)
CarsXE MCP — wraps the CarsXE Vehicle Data API (carsxe.com)
Vehicle data for AI: VIN decoder, automotive specs, stolen checks, valuation and way more.
NHTSA MCP — wraps the NHTSA vPIC (Vehicle Product Information Catalog) API (free, no auth)
