Fusion 360 MCP

# FusionMCP Quick Start Guide Get up and running with FusionMCP in 5 minutes! ## 📋 Prerequisites - Python 3.11+ - Fusion 360 (2025) - At least one AI provider (Ollama recommended for beginners) ## 🚀 5-Minute Setup ### Step 1: Install Ollama (Easiest Option) ```bash # macOS/Linux curl -fsSL https://ollama.com/install.sh | sh # Or download from https://ollama.com # Pull a model ollama pull llama3 ``` ### Step 2: Clone and Install ```bash # Clone repository git clone https://github.com/yourusername/fusion360-mcp.git cd fusion360-mcp # Create virtual environment python -m venv venv source venv/bin/activate # Windows: venv\Scripts\activate # Install dependencies pip install -r requirements.txt ``` ### Step 3: Configure ```bash # Copy example config cp examples/example_config.json config.json # Edit config.json - for Ollama only, just verify: # "default_model": "ollama:llama3" ``` ### Step 4: Start Server ```bash # Option 1: Use helper script ./start_server.sh # or start_server.bat on Windows # Option 2: Direct command python -m mcp_server.server ``` You should see: ``` ✅ Logger initialized with level INFO ✅ Initialized MCP Router with providers: ['ollama'] ✅ MCP Server started on 127.0.0.1:9000 ``` ### Step 5: Install Fusion Add-in **macOS**: ```bash cp -r fusion_addin ~/Library/Application\ Support/Autodesk/Autodesk\ Fusion\ 360/API/AddIns/FusionMCP ``` **Windows**: ```bash xcopy /E /I fusion_addin "%APPDATA%\Autodesk\Autodesk Fusion 360\API\AddIns\FusionMCP" ``` ### Step 6: Test in Fusion 360 1. Open/Restart Fusion 360 2. Go to **Tools** → **Add-Ins** → **Scripts and Add-Ins** 3. Select **FusionMCP** → Click **Run** 4. Click the **MCP Assistant** button in toolbar 5. Try these commands: ``` Create a 20mm cube Design a mounting bracket with 4 holes Make a cylindrical shaft 10mm diameter, 50mm long ``` ## ✅ Verify Installation ### Test Server (Terminal) ```bash # Health check curl http://127.0.0.1:9000/health # Test command curl -X POST http://127.0.0.1:9000/mcp/command \ -H "Content-Type: application/json" \ -d '{ "command": "ask_model", "params": { "provider": "ollama", "model": "llama3", "prompt": "Create a 10mm cube" } }' ``` Expected response: ```json { "status": "success", "actions_to_execute": [ { "action": "create_box", "params": {"width": 10, "height": 10, "depth": 10, "unit": "mm"} } ] } ``` ## 🔧 Adding More Providers ### OpenAI (Recommended for Production) 1. Get API key from https://platform.openai.com 2. Edit `config.json`: ```json { "openai_api_key": "sk-proj-YOUR_KEY_HERE", "default_model": "openai:gpt-4o-mini" } ``` 3. Restart server ### Google Gemini (Fast & Affordable) 1. Get API key from https://makersuite.google.com/app/apikey 2. Edit `config.json`: ```json { "gemini_api_key": "AIza_YOUR_KEY_HERE", "default_model": "gemini:gemini-1.5-flash-latest" } ``` ### Claude (Best Reasoning) 1. Get API key from https://console.anthropic.com 2. Edit `config.json`: ```json { "claude_api_key": "sk-ant-YOUR_KEY_HERE", "default_model": "claude:claude-3-5-sonnet-20241022" } ``` ## 🎯 Example Workflows ### Simple Geometry ``` User: "Create a 50mm cube" → Creates 50x50x50mm box ``` ### Complex Design ``` User: "Design a phone stand" → AI asks clarifying questions about dimensions → Generates multi-step design with base, support, and angle ``` ### Parametric Design ``` User: "Make a mounting bracket 100x50mm with 4 M5 holes at corners" → Creates base plate 100x50x5mm → Adds 4x 5.5mm holes with 10mm edge offset ``` ## 🐛 Common Issues ### Issue: "Connection refused" **Solution**: Make sure MCP server is running (`python -m mcp_server.server`) ### Issue: "Ollama not found" **Solution**: Install Ollama and verify with `ollama list` ### Issue: "Add-in not visible in Fusion" **Solution**: - Check folder name is exactly `FusionMCP` - Restart Fusion 360 - Verify path: Scripts and Add-Ins → Add-Ins tab ### Issue: "API key invalid" **Solution**: - Verify key in config.json - Check no extra spaces or quotes - Try .env file instead ## 📚 Next Steps 1. **Read Full Docs**: See [README.md](README.md) for complete documentation 2. **Explore Examples**: Check `examples/` folder for more use cases 3. **Customize System Prompt**: Edit `prompts/system_prompt.md` for behavior changes 4. **Join Community**: Star on GitHub, join discussions ## 💡 Pro Tips 1. **Use Fallback Chain** for reliability: ```json "fallback_chain": [ "openai:gpt-4o-mini", "gemini:gemini-1.5-flash-latest", "ollama:llama3" ] ``` 2. **Enable Caching** for conversation history: ```json "cache_enabled": true, "cache_type": "json" ``` 3. **Adjust Temperature** for precision: - `0.3-0.5`: Precise, predictable (engineering) - `0.7-0.9`: Creative, varied (design exploration) 4. **Check Logs** when debugging: ```bash tail -f logs/mcp_server.log ``` ## 🎉 Success! You're now ready to use AI-powered parametric CAD design! Try asking: *"Design a custom enclosure for a 50x30x20mm PCB with mounting tabs"* --- **Need Help?** Open an issue on GitHub or check the troubleshooting guide in README.md