MayaMCP

MayaMCP is a server that exposes a collection of tools to control Autodesk Maya via the Model Context Protocol. It includes basic scene operations like listing objects by type, creating objects, getting and setting object attributes, and managing scenes (new, open, save) as well as selecting objects. Advanced tools enable creating complex models, performing mesh operations, managing materials and curves, and generating complete scenes. Clients can invoke these tools through the MCP interface, receive structured results, and iterate with natural language prompts to drive Maya workflows. To use MayaMCP, run the Python server script, then configure your MCP client (e.g., Claude Desktop) to point at the local server and load the MayaMCP toolset. Once connected, you’ll see a catalog of available tools and can start issuing commands like creating a cube, listing all cameras, or applying a material to an object. The server is designed to be easily extensible by adding new tool files in the Maya environment, making it simple to grow capabilities over time.

Prerequisites

• Python 3.10 or greater
• Git
• Access to Maya for testing (optional, for validating tool operations)

1. Clone the repository

• git clone https://github.com/your-org/mayamcp.git
• cd mayamcp

2. Create a Python virtual environment

• python -m venv .venv

3. Activate the virtual environment

• Windows: ..venv\Scripts\activate.bat
• macOS/Linux: source .venv/bin/activate

4. Install dependencies

• pip install -r requirements.txt

5. Run the MCP server

• Use the command configuration below to run the server via your MCP client setup.
• Example (replacing path with your local repo path):
  • python [FULL PATH TO MayaMCP PROJECT]/src/maya_mcp_server.py

6. Configure your MCP client

• In your MCP client (e.g., Claude Desktop), add an MCP server entry pointing to the above command. For example: { "mcpServers": { "MayaMCP": { "command": "python", "args": ["[FULL PATH TO MayaMCP PROJECT]/src/maya_mcp_server.py"] } } }

Notes

• Ensure Maya is accessible if you intend to run actual Maya commands; the server sends Python snippets to Maya via the command port.
• The server relies on Python 3.10+ and is designed to be extended by adding tool scripts in the mayatools/thirdparty directory.

Tips and common issues:

• Use a dedicated virtual environment to avoid conflicts with system Python.
• If you encounter connection or port issues, verify that the command port used by Maya is accessible and that Maya is configured to allow external scripting connections.
• Tools are dynamically loaded; to add new functionality, drop a new tool Python file into the mayatools/thirdparty directory and restart the server.
• When testing, run Maya and the MCP server side-by-side to validate end-to-end interactions. The server captures tool signatures dynamically, so well-documented function names and docstrings help MCP clients present meaningful prompts to users.
• If you modify tools, ensure proper error handling so the MCP client can surface helpful error messages to users.