by gNucleus
Generate editable CAD parts or assemblies directly from natural‑language prompts by leveraging gNucleus generative AI models through a lightweight MCP server.
The server provides a Model Context Protocol (MCP) endpoint that accepts a single text input and returns a CAD model generated by gNucleus’s AI. It works with any MCP‑compatible client (e.g., Claude Desktop, custom scripts) to turn descriptions like “draw a block with length=80mm” into a 3‑D model hosted on gNucleus.
.env
file with GNUCLEUS_HOST
and GNUCLEUS_API_KEY
(enterprise users also add GNUCLEUS_ORG_ID
).pip install -r requirements.txt
.python main.py
or test it using the MCP inspector: npx @modelcontextprotocol/inspector python3 main.py
.text_to_cad
tool.text_to_cad(input: str)
Q: Which Python version is required?
A: Python 3.7 or newer.
Q: Do I need a gNucleus account?
A: Yes. Sign up at https://gnucleus.ai and create an API key (personal or team).
Q: How is authentication handled?
A: The server reads GNUCLEUS_API_KEY
from the .env
file and passes it to the gNucleus API.
Q: Can I download the generated CAD file?
A: Direct download isn’t supported yet. Open the shared viewer URL, log in to gNucleus, and download the model from there.
Q: What if the model looks incomplete?
A: Try the same prompt on the gNucleus web UI and download the file to inspect it; often the issue is a partial generation that can be resolved by re‑prompting.
Q: How do I run the server on a remote machine?
A: Ensure the .env
file is secured, expose the desired port, and start python main.py
in a background process or container.
gNucleus is a Generative AI platform that transforms text and images into editable CAD models.
This repository contains the gNucleus Text-To-CAD MCP server, which connects to the gNucleus API. It allows MCP clients to generate CAD parts or assemblies from text input using gNucleus’s GenAI models.
Clone this repository
Create and activate a virtual environment (recommended):
python -m venv .venv
source .venv/bin/activate # On Windows: .venv\Scripts\activate
Install dependencies:
pip install -r requirements.txt
Create a .env
file in the root directory with the following variables:
Invidisual User
GNUCLEUS_HOST="genai.gnucleus.ai"
GNUCLEUS_API_KEY=<your-personal-api-key>
Enterprise User
GNUCLEUS_HOST="genai.gnucleus.ai"
GNUCLEUS_API_KEY=<your-team-key>
GNUCLEUS_ORG_ID=<your-team-id>
Start the MCP server:
python main.py
You can test the MCP server using the inspector by running
npx @modelcontextprotocol/inspector python3 main.py
The following MCP tools are available:
When used with LLMs that support the MCP protocol, this server enables natural language interaction with gNucleus:
i18n support: You can use any language as the prompt, in general english works better than other languages, but it should work in general.
The generated result include the design spec with Key Parameter, Description and a shared URL with 3D display viewer. The shared URL will be expired in 24 hrs.
If you use virtual python enviroment, use this config:
{
"mcpServers": {
"gnucleus": {
"command": "/ABSOLUTE/PATH/TO/YOUR/text-to-cad-mcp/.venv/bin/python",
"args": [
"/ABSOLUTE/PATH/TO/YOUR/text-to-cad-mcp/main.py"
],
"workingDirectory": "/ABSOLUTE/PATH/TO/YOUR/text-to-cad-mcp",
"env": {
"GNUCLEUS_HOST": "genai.gnucleus.ai",
"GNUCLEUS_API_KEY": "YOUR_API_KEY_HERE",
}
}
}
}
If you use global python enviroment and also installed the requirments.txt into your global python enviroment, use this config
{
"mcpServers": {
"gnucleus": {
"command": "python",
"args": [
"/ABSOLUTE/PATH/TO/YOUR/text-to-cad-mcp/main.py"
],
"workingDirectory": "/ABSOLUTE/PATH/TO/YOUR/text-to-cad-mcp",
"env": {
"GNUCLEUS_HOST": "genai.gnucleus.ai",
"GNUCLEUS_API_KEY": "YOUR_API_KEY_HERE",
}
}
}
}
You should see the gNucleus Text-To-CAD tools in Claude
Chat with Claude and add "in CAD" in each prompt or tell Claude to use "text-to-cad tools"can help Claude to trigger the Text-To-CAD tool better.
The generated result include the design spec with Key Parameter, Description and a shared link with 3D display viewer. The shared URL will be expired in 24 hrs. If Claude didn't output the shared URL in the chat message, you can ask Claude to always output the shared URL.
Click the shared URL, it will display the CAD model in gNucleus 3D viewer, you can
Example 1: Text To CAD Part
Input and CAD Part design spec
CAD Part in 3D viewer
Example 2: Text To CAD Assembly
Input and CAD Assembly design spec
CAD Assembly in 3D viewer
Note: Downloading CAD models from the viewer is not yet supported. To download the model, please log in to https://gnucleus.ai and try the same prompt using the full feature set.
If the CAD model looks incorrect or only partially generated, try logging in at https://gnucleus.ai, enter the same prompt, and download the CAD file (e.g., FreeCAD format). Then open it in your CAD software(e.g. FreeCAD ). This issue is often caused by incomplete generation of CAD features within the part.
You can also report a bug or contact us with the model link and prompt at https://gnucleus.ai/contact.
.env
file and never commit it to githubPlease log in to share your review and rating for this MCP.
{ "mcpServers": { "gnucleus": { "command": "npx", "args": [ "-y", "@modelcontextprotocol/inspector", "python3", "main.py" ], "env": { "GNUCLEUS_HOST": "genai.gnucleus.ai", "GNUCLEUS_API_KEY": "<YOUR_API_KEY>" } } } }
Explore related MCPs that share similar capabilities and solve comparable challenges
by sunsetcoder
Track flights in real time, retrieve arrival and departure times, view airport status, and monitor emergency situations using Flightradar24 data via a Model Context Protocol server.
by EduBase
Enables Claude and other LLMs to interact with the EduBase e‑learning platform via the Model Context Protocol, allowing automated quiz creation, exam scheduling, and result analysis.
by jyjune
Provides control and access to a video surveillance system, allowing retrieval of live and recorded video streams, channel status queries, PTZ camera positioning, and playback dialog management.
by NonicaTeam
Enables AI desktop applications to read and select Revit model elements through a set of predefined micro‑tools, integrated directly into the Nonicatab toolbar.
by wuye-ai
Provides Model Context Protocol endpoints for property‑industry AI services, including daily industry news, knowledge‑base search, and intelligent question‑answering.
by zaiwork
MCP Server for ZIZAI Recruitment API.
by AndrewKlement
Provides real-time access to Gaggiuino espresso machine status, latest shot ID, and detailed shot telemetry for AI clients.
by zed-industries
A high‑performance, multiplayer code editor designed for speed and collaboration.
by modelcontextprotocol
A Model Context Protocol server that provides web content fetching capabilities.