Fusion

Connects LLMs with Autodesk Fusion to execute CAD actions via MCP with code execution, screenshots, and parameter control.
  • typescript

1

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "shiguri-01-fusion-mcp": {
      "command": "uvx",
      "args": [
        "--from",
        "[absolute_path_to_repo]/fusion-mcp/mcp-server",
        "fusion-mcp-server"
      ]
    }
  }
}

Fusion MCP lets you connect an LLM with Autodesk Fusion to drive CAD operations through natural language. It enables a conversational workflow where your AI partner can request Fusion actions and execute them via an MCP server and add-in.

How to use

You use an MCP client to access the Fusion MCP server and its add-in. Start the MCP server locally, register it with your client, and then issue high‑level CAD prompts. The MCP server exposes tools that let the LLM run Python code inside Fusion, grab screenshots of the active viewport, and manage User Parameters. For best results, provide concrete values and clear step-by-step instructions to the LLM so it can sequence actions like selecting parts and performing operations.

How to install

Prerequisites you need to have installed before proceeding:

  • Autodesk Fusion
  • an MCP client (for example Claude Desktop)
  • Python
  • uv

Follow these concrete steps to set up the Fusion MCP server and add-in.

  1. Download the Fusion MCP setup files.

  2. Register the MCP server with your MCP client. In Claude Desktop, open the developer configuration and add the MCP server configuration as shown below.

{
  "mcpServers": {
    "fusion": {
      "command": "uvx",
      "args": [
        "--from",
        "[absolute_path_to_repo]/fusion-mcp/mcp-server",
        "fusion-mcp-server"
      ]
    }
  }
}
  1. Register the Fusion add-in in Fusion.

a) In Fusion, go to UTILITIES → ADD-INS to open Scripts and Add-Ins window.

b) Click the + button → Script or add-in from device and select the downloaded fusion-mcp/mcp-addin folder.

c) The mcp-addin appears in the list. Click Run to start it. Enable Run on Startup if you want the add-in to start automatically when Fusion launches.

Additional setup notes

Security: keep in mind that the Fusion add-in can execute arbitrary Python code generated by the LLM inside Fusion. Create a known-good backup of Fusion data before using the system, and review any generated code before running it.

Available tools

execute_code

Executes Python code through the Fusion API to perform CAD operations inside Fusion.

get_viewport_screenshot

Captures a screenshot of the active Fusion viewport for visual verification.

list_user_parameters

Retrieves the list of User Parameters defined in the Fusion model.

set_user_parameter

Sets the value of a User Parameter using an expression or direct value.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
Fusion MCP Server - shiguri-01/fusion-mcp | VeilStrat