dcc-mcp/dcc-mcp-maya

View on GitHub ↗

Maya-specific implementation of Model Context Protocol (MCP) for Autodesk Maya, enabling remote control and integration with MCP ecosystem.

43 ★5 forksPythonUpdated 21d ago

What you need to know

Maya domain type adapter for the dcc-mcp ecosystem, turning Autodesk Maya into a standards-compliant MCP Streamable HTTP backend. Agents control Maya through the shared dcc-mcp-cli gateway or direct MCP endpoint, with typed skills for scene, geometry, materials, animation, rigging, rendering and export workflows.

Install

mayapy -m pip install dcc-mcp-maya

Usage

  • dcc-mcp-cli dcc-types; dcc-mcp-cli list
  • dcc-mcp-cli search --query '<task>' --dcc-type maya
  • dcc-mcp-cli describe <tool-slug>
  • dcc-mcp-cli call <tool-slug> --json '{"key":"value"}'
  • Load plugin from Window > Settings/Preferences > Plug-in Manager, then point MCP host at http://127.0.0.1:9765/mcp

Key features

  • 76+ discoverable typed Maya tools across 28 bundled skill packages
  • Progressive loading with compact boot surface (maya-scripting + core maya-scene)
  • Rust dcc-mcp-server sidecar keeps HTTP/gateway traffic out of Maya's UI thread with Qt dispatcher bridge
  • Multi-instance gateway with optional LAN exposure
  • Operational guardrails: readiness probes, project persistence, MCP resources, job persistence, crash cleanup, cancellation, script-execution lockouts
  • DCC UI Control-style capabilities via the core framework

Best for

VFX/3D studios and TDs that want agents to drive Autodesk Maya scene, mesh, render and pipeline work

Caveats

  • Requires Autodesk Maya 2020+ and Python 3.7+
  • DCC_MCP_MAYA_DISABLE_EXECUTE_PYTHON/MEL can block arbitrary execution for production
Platforms: Windows · macOS · LinuxClients: MCP Streamable HTTP clients · dcc-mcp-cli
Documentation ↗

Reviewed 2026-08-11

Topics

ai-agentsautodeskdccdcc-mcpmayamcpmodel-context-protocolpython
Stars
43★
Forks
5
Language
Python
License
MIT
Created
2025-03-18
Last push
2026-08-14