fusion-mcp
Open-source MCP server for controlling Autodesk Fusion 360 from any AI agent, enabling 3D modeling operations via natural language.
README
fusion-mcp
Open-source MCP server for controlling Autodesk Fusion 360 from any AI agent. just Python.
MCP Client ──stdio──▶ MCP Server (Python) ──TCP──▶ Fusion 360 Add-in ──▶ Fusion API
Setup
1. Clone and install
With uv (recommended):
git clone https://github.com/prim-design/fusion-mcp.git
cd fusion-mcp
uv pip install -e .
With pip (macOS/Linux):
git clone https://github.com/prim-design/fusion-mcp.git
cd fusion-mcp
python3 -m venv .venv
source .venv/bin/activate
pip install -e .
With pip (Windows):
git clone https://github.com/prim-design/fusion-mcp.git
cd fusion-mcp
python -m venv .venv
.venv\Scripts\activate
pip install -e .
Requires Python 3.10+.
2. Install the Fusion 360 add-in
Copy the addin/ folder to Fusion's add-ins directory:
macOS:
cp -r addin/ "$HOME/Library/Application Support/Autodesk/Autodesk Fusion 360/API/AddIns/FusionMCP"
Windows:
xcopy /E addin "%APPDATA%\Autodesk\Autodesk Fusion 360\API\AddIns\FusionMCP\"
Then in Fusion 360: Shift+S → Add-Ins → select FusionMCP → Run (optionally check "Run on Startup").
3. Connect your MCP client
Claude Code:
claude mcp add fusion-mcp -- /path/to/fusion-mcp/.venv/bin/fusion-mcp
Claude Desktop — add to claude_desktop_config.json:
{
"mcpServers": {
"fusion-mcp": {
"command": "/path/to/fusion-mcp/.venv/bin/fusion-mcp"
}
}
}
Works with any MCP client — run .venv/bin/fusion-mcp as a stdio server.
Tools (51)
| Category | Tools |
|---|---|
| Sketch | create_sketch finish_sketch draw_line draw_rectangle draw_circle draw_arc draw_polygon draw_spline draw_slot |
| 3D Features | extrude revolve sweep loft |
| Modifications | fillet chamfer shell draft |
| Patterns | pattern_rectangular pattern_circular mirror |
| Booleans | combine split_body |
| Components | create_component list_components delete_component move_component rotate_component rename |
| Joints | create_joint (all 7 types) set_joint_limits drive_joint list_joints |
| Rigid Groups | create_rigid_group list_rigid_groups delete_rigid_group |
| Inspection | get_design_info get_body_info measure check_interference fit_view screenshot |
| Export/Import | export_stl export_step export_3mf import_mesh |
| Utility | undo delete_body delete_sketch batch |
| Escape Hatch | execute_python — run arbitrary Python inside Fusion with full API access |
Joint types
create_joint supports all 7 Fusion joint types: rigid, revolute, slider, cylindrical, pin_slot, planar, ball. Uses as-built joints by default (components stay in place).
Screenshot
screenshot captures the viewport as a PNG image returned directly to the AI. Supports preset views (front, back, top, bottom, left, right, iso) and custom camera positions via eye_x/y/z + target_x/y/z.
Important notes
- Units: All lengths in cm, all angles in degrees
- Z-negation gotcha: On XZ plane, sketch +Y = world -Z. On YZ plane, sketch +X = world -Z.
- Thread safety: The add-in dispatches all API calls to Fusion's main thread via CustomEvent
- Escape hatch:
execute_pythonruns arbitrary Python inside Fusion for anything not covered by curated tools
How it works
The MCP server (server/) runs as a stdio process your AI agent connects to. It forwards commands over a TCP socket to the Fusion 360 add-in (addin/), which runs inside Fusion and executes API calls on the main thread using Fusion's CustomEvent system for thread safety.
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.