dcc-mcp

dcc-mcp-blender: Control Blender Scenes Through MCP

dcc-mcp-blender embeds a Streamable HTTP MCP server in Blender, built on Blender’s Python API and the shared dcc-mcp-core runtime. It gives AI agents typed tools for modeling, UVs, rigging, materials, animation, and rendering, with support for Blender 4.5 and 5.2.

ActivePythonMITMCPAI AgentsAutomation
Stars
59
Forks
11
Stars, last 30 days
+17
Last commit
8 days ago
Contributors
5

Our take

Activeregular commits and releases

Health score

66/100

How it is scored
Activity
86
Community
21
Issues
74
Pull requests
83

+17 stars in the last 30 days

  • Merges more pull requests than 95% of the projects we track
  • Releases more often than 89% of the projects we track
  • Closes issues faster than 81% of the projects we track

This is a feature-rich Blender automation adapter with active development and releases, but adoption carries contributor-concentration risk because most contributions come from one person.

Good fit if

  • You need AI-agent control of Blender through typed MCP tools across modeling, look development, animation, or rendering.
  • You use Blender 4.5 or 5.2 and want either an in-process add-on or headless automation.
  • You want an MIT-licensed project that has maintained a steady commit and release cadence.

Look elsewhere if

  • You need support for Blender versions older than 4.5, which the project says it does not support.
  • Your team cannot accept dependence on a project where contributions are highly concentrated in one contributor.
  • You need a narrow, lightweight Blender integration rather than a broad tool and skill catalog.
All health signals
Last commit2026-10-25 (8 days ago)
Commits, last 90 days100+
Releases, last 12 months68 (latest v0.2.26, 2026-10-25)
Contributors5 (top contributor: 95% of commits)
Issues closed, last 90 days3+ (typically closed in 0 days)
Pull requests merged, last 90 days93 (typically merged in 0 days)
Project age7 months

Checked on 2026-11-03 with the GitHub API.

Project history

Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.

Selected snapshot

Stars

64

Forks

14

Change since Nov 17, 2026: +1 stars·+1 forks

Overview

dcc-mcp-blender embeds a Streamable HTTP MCP server in Blender, built on Blender’s Python API and the shared dcc-mcp-core runtime. It lets MCP-compatible AI agents work with editable Blender scenes through typed tools rather than relying only on generated scripts. The README documents Blender 4.5 LTS and 5.2 as CI targets, with Windows, Linux, and macOS packaging.

The project targets artists and technical teams who want agent-driven 3D work integrated with Blender and, through the DCC-MCP ecosystem, other DCC applications. Its broad tool catalog spans scene and mesh editing, UVs, rigging, materials, animation, simulation, interchange, validation, and rendering. A Python API and headless bootstrap also support scripts and CI.

Key Features

  • Runs its MCP server inside Blender and communicates over Streamable HTTP.
  • Provides typed tools for scene management, object and mesh operations, UV mapping, rigging, and animation.
  • Supports materials, shader nodes, compositor graphs, lighting, rendering, and texture baking.
  • Includes import, export, asset validation, publishing, and render-job tools.
  • Offers a skill system for staged tool loading and optional semantic skill recall.
  • Supports GUI add-on use, a Python API, and headless Blender automation.
  • Can disable arbitrary Python and script execution through environment settings.

Use Cases

  • A 3D artist can ask an MCP-compatible agent to build or revise editable Blender geometry, materials, and lighting.
  • A technical artist can automate repeated rigging, UV, animation, or asset-validation tasks with typed operations.
  • A pipeline team can run Blender headlessly in CI to prepare, validate, or render assets.
  • A team using several DCC applications can use the shared DCC-MCP gateway and CLI to coordinate Blender with other adapters.

What you need

Detected in the repository

  • Python >=3.7 (from pyproject.toml)
  • A test suite and automated checks on GitHub Actions

License in plain words

MITpermissive

  • Commercial use: yes
  • Modify and redistribute: yes
  • You must keep: the copyright and license notice
  • Share your changes: no

A summary, not legal advice: the LICENSE file is what applies.

Getting Started

For a script or CI installation, install the package:

pip install dcc-mcp-blender

Then start the server from Blender's Python environment:

import dcc_mcp_blender
dcc_mcp_blender.start_server()

For the Extension ZIP installation, agent setup, and gateway configuration, see the README.

Considerations

The documented minimum host is Blender 4.5, and the project’s CI targets Blender 4.5 LTS and 5.2. The release ZIP is intended for Blender’s Extensions installation flow; the legacy add-on installation route is explicitly unsupported. GUI and GPU workflow acceptance is not established by the documented background tests, so teams should validate their own interactive or hardware-dependent workflows. The project is actively maintained, but its contributor base is small and highly concentrated, which may affect continuity if the dominant contributor steps away.

Found this useful?

Share it with someone who would like dcc-mcp-blender.