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., "@Hardware MCP Server design a heatsink for a 50W CPU and simulate its thermal performance"
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.
This repository shows examples of how you can use a language model to perform tool calls for:
building a CAD part
running heat transfer simulations
This repository uses the Model Context Protocol (MCP)
First, clone the current GitHub repository with the following command.
git clone https://github.com/TheFloatingString/mcp-for-hardware.git
cd mcp-for-hardwareNow, install all the Python dependencies using the uv package manager:
pip install uv
uv syncInstall ngrok or any similar tunelling services. ngrok can be installed from https://ngrok.com/downloads
Next, if you use ngrok, run:
ngrok http 8000This will expose port 8000 (which we will be using soon for the MCP server) to a specific URL.
You should see a publicly visible URL after running the ngrok command. Keep note of this, you will need to add this in the .env file's environment variables.
Create a .env file and add the following environment variables
SERVER_BASE_URL=""
ANTHROPIC_API_KEY=""In one terminal, run the MCP server with the following:
uv run server.pyIn a second terminal, run the client with the following:
uv run client.pyThis server cannot be deployed
Parametric 3D CAD for AI agents: build print-ready parts, check them, export STL, 3MF or STEP.
AI-callable calculators and engineering models with real formulas. No hallucinated math.
Design, solve and simulate HVAC systems from real components, weather years and buildings.
Agent-first CAD: editable .kcad.ts source, deterministic review, OpenCASCADE kernel.
