# Godot MCP Server A Model Context Protocol (MCP) server that enables Claude Desktop to control and interact with the Godot Engine editor. # DEMO VIDEO https://github.com/user-attachments/assets/07424399-31b5-47ee-a20d-808b2e789731 # NEW UPDATE!!!! ADDED MESHY API INTEGRATION Screenshot 2025-07-14 at 9 07 13 PM GENERATE DYNAMIC SCENES BY CALLING THE MESHY API, DIRECTLY IMPORTED INTO GODOT ## Setup Instructions ### Prerequisites - Godot Engine (4.x or later) - Python 3.8+ - Claude Desktop app - Meshy API account (optional, for AI-generated meshes) ### STEP 0: Clone the repo and navigate to the directory ```bash git clone https://github.com/Dokujaa/Godot-MCP.git ``` ### Step 1: Install Godot Plugin 1. Copy the `addons/godot_mcp/` folder to your Godot project's `addons/` directory 2. Open your Godot project 3. Go to `Project → Project Settings → Plugins` 4. Enable the "Godot MCP" plugin 5. You should see an "MCP" panel appear at the bottom of the editor 6. The plugin automatically starts listening on a port ### Step 2: Set up Python Environment 1. Navigate to the `python/` directory: ```bash cd python ``` 2. Create and activate a virtual environment: ```bash python3 -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate ``` 3. Install dependencies: ```bash pip install -r ../requirements.txt ``` ### Step 3: Configure Claude Desktop 1. Locate your Claude Desktop configuration file: - **macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json` - **Windows**: `%APPDATA%\Claude\claude_desktop_config.json` 2. Add the Godot MCP server configuration: ```json { "mcpServers": { "godot": { "command": "/path/to/your/godot-mcp/python/venv/bin/python", "args": ["/path/to/your/godot-mcp/python/server.py"], "env": {} } } } ``` Replace `/path/to/your/godot-mcp/python/server.py` with the actual path to your server.py file. 3. Restart Claude Desktop and happy prompting! ### OPTIONAL: Set up Meshy API 1. Sign up for a Meshy API account at [https://www.meshy.ai/](https://www.meshy.ai/) 2. Get your API key from the dashboard (format: `msy-`) 3. Set up your API key using one of these methods: **Option A: Using .env file (Recommended)** ```bash # Copy the example file cp python/.env.example python/.env # Edit the .env file and add your API key nano python/.env # or use your preferred editor ``` Then add your key to the `.env` file: ``` MESHY_API_KEY=your_actual_api_key_here ```