Skip to main content
Glama
Maya MCP
by azagoruyko

Maya MCP

Maya MCP connects an AI client to an open Autodesk Maya scene. It has two parts: an external MCP server and a small bridge running inside Maya.

Requirements

  • Python 3.10 or newer for the external MCP server.

  • Autodesk Maya with pyzmq installed in Maya's Python environment.

  • Rig Builder and RigSystem available in Maya for reusable, module-based rig workflows. Scene inspection and direct MCP tools can be used without them.

Related MCP server: maya-mcp

Setup

Install the MCP server in a regular Python environment outside Maya:

cd D:\path\to\maya_mcp
py -m venv .venv
.\.venv\Scripts\python.exe -m pip install -e .

In your MCP client, configure the server to run:

D:\path\to\maya_mcp\.venv\Scripts\python.exe -m maya_mcp.mcp_server

In Maya's Python Script Editor, start the bridge:

import sys

project_path = r"D:\path\to\maya_mcp"
if project_path not in sys.path:
    sys.path.insert(0, project_path)

import maya_mcp
maya_mcp.start()

The default local endpoint is tcp://127.0.0.1:51608. Set the same MAYA_MCP_ENDPOINT in both environments to change it. Optionally, set the same MAYA_MCP_TOKEN in both environments to protect the connection. Stop the bridge from Maya with maya_mcp.stop().

Using Maya MCP

The server provides these tools:

  • scene_context — current scene, selection, roots, units, references, and counts.

  • scene_naming — naming patterns and likely control families.

  • scene_hierarchy — a focused DAG hierarchy.

  • scene_connections — connections between nodes and attributes.

  • viewport_capture — an image of the Maya viewport.

  • maya_command — a low-level fallback for native maya.cmds calls.

For rig edits, the intended workflow is to find or compose modules in Rig Builder and use RigSystem's high-level APIs from those modules. Inspect and verify the scene with Maya MCP before and after running a workflow. Do not guess ambiguous inputs such as edge loops or guide curves; provide them as module inputs and run the module after they are specified.

AI workflow guidance is included in the server instructions and available as the maya://ai-workflow MCP resource.

Available Tools

6 tools
maya_commandMaya CommandC

Execute one native maya.cmds command or a batch in a single undo chunk.

Inputs: args, kwargs, command, commands. Returns: object

scene_connectionsScene ConnectionsC

Follow compact input/output plug connections from nodes or plugs.

Inputs: depth, targets, direction, exclude_types, +2 more. Returns: object

scene_contextScene ContextC

Get scene metadata, selection, roots, units, time, and node counts.

Inputs: root_limit, count_types, include_plugins. Returns: object

scene_hierarchyScene HierarchyC

Read a compact text DAG hierarchy from roots or the current selection.

Inputs: depth, roots, max_nodes, include_types, +1 more. Returns: object

scene_namingScene NamingC

Summarize common prefixes, suffixes, families, and NURBS control candidates.

Inputs: family_limit, max_examples, prefix_limit, suffix_limit. Returns: object

viewport_captureViewport CaptureC

Capture the active Maya viewport or a specified camera and return it as a PNG image.

Inputs: panel, width, camera, height, +2 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    B3.2/5.0

    Scored across 6 tools

    Disambiguation4/5

    The four scene_* inspection tools have distinct targets: metadata, naming patterns, hierarchy, and connections. The only ambiguity comes from maya_command, a generic escape hatch that can overlap with every specialized tool, though its intended use as a native-command executor is clear.

    Naming Consistency5/5

    All names use consistent two-word snake_case with clear domain prefixes such as scene_ and viewport_. The convention is predictable and readable even though it is noun-oriented rather than verb_noun.

    Tool Count5/5

    Six tools is a well-scoped set for Maya scene inspection plus command execution. Each tool has a focused role, and there is no bloat or missing obvious specialization.

    Completeness4/5

    The set covers key read operations, viewport capture, and arbitrary command execution, so agents are not blocked. It lacks dedicated structured tools for common write or attribute-editing workflows, but maya_command provides a workaround.

    Related MCP Connectors

    Related MCP Servers

    • F
      license
      Not graded
      quality
      D
      maintenance
      Enables AI assistants to programmatically control Autodesk Maya via natural language using over 30 tools for 3D modeling, lighting, and animation. It connects through Maya's command port to facilitate procedural scene generation and complex production-ready workflows.
      1
      -
    • A
      license
      B
      quality
      C
      maintenance
      Enables AI agents to drive Autodesk Maya through semantic tools covering scene, object, material, animation, and rendering operations, paired with execute_python/execute_mel escape hatches and read-only resources. Also supports multi-instance routing, viewport capture for visual iteration, and asynchronous render job polling.
      22
      1
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables natural language control of Autodesk Maya through MCP, including scene queries, Python/MEL execution, viewport screenshots, tool invocation, and customizable body rigging.
      MIT
    • A
      license
      A
      quality
      B
      maintenance
      Gives LLM agents spatial awareness inside Autodesk Maya, enabling scene inspection and measurement, aesthetic and engineering audits, camera planning, checkpoint/rollback safety, asset import, and viewport or render previews. Agents can also execute arbitrary Python in Maya to build and verify 3D scenes against engineering rules instead of writing blind.
      25
      76 PyPI
      5
      MIT