An open-source autonomous research agent that plans questions, gathers web and local sources, and writes cited reports with any LLM provider.
ahujasid/mcp-for-blender
Blender add-on and MCP server that let AI assistants like Claude build, modify and inspect 3D scenes in Blender through prompts.
- Stars
- 31,634
- Forks
- 2,971
- Open issues
- 52
- Last commit
- 2 days ago
What is it?
MCP for Blender connects Blender to any large language model that supports the Model Context Protocol. It has two parts: a Blender add-on that runs a socket server inside Blender to receive and execute commands, and a Python MCP server that exposes those commands to AI clients such as Claude Desktop, Claude Code, Codex, Cursor and VS Code.
Once connected, the AI can write Blender Python to create and edit objects, apply materials, lay out scenes and handle animation, rigging and retopology. It can also look at its own output through viewport, camera, multi-angle and render views. Scene inspection, Poly Haven, Sketchfab and Poly Pizza asset search and import, and AI model generation through Tripo, Hyper3D Rodin, Hunyuan3D and Meshy are included.
A safe mode can be enabled to check scripts before they run and block file, network and process access. The project is an unofficial community plugin and is not made by the Blender Foundation.
Key features
- 01Create, modify and delete 3D objects and apply materials through AI prompts
- 02Run Blender Python code directly in a live Blender session
- 03Check results with viewport, camera, multi-angle, wireframe and render views
- 04Search and import assets from Poly Haven, Sketchfab and Poly Pizza
- 05Generate 3D models with Tripo, Hyper3D Rodin, Hunyuan3D and Meshy
- 06Optional safe mode that screens scripts before execution
Use cases
- Prototyping a 3D scene by describing it in natural language
- Populating a scene with HDRIs, textures and low-poly assets from free libraries
- Automating repetitive Blender modeling or layout tasks with an AI assistant
- Generating a starting 3D model from a text prompt or reference image
Quick start
- 1
Run the one-line installer (macOS/Linux)
Install with the official instructions in the README - 2
Or, if uv is already installed, run setup
uvx mcp-for-blender setup - 3
Fully quit and reopen your AI app, then open Blender; the addon starts its server automatically
Things to know
- Blender must be running with the addon enabled for the AI to control it.
- The addon's socket server has no authentication or encryption, so keep it on localhost unless on a trusted network.
- Some asset and AI model generation features need third-party API keys or a paid Premium tier.
Topics
FAQ
- What is mcp-for-blender?
- Blender add-on and MCP server that let AI assistants like Claude build, modify and inspect 3D scenes in Blender through prompts. MCP for Blender connects Blender to any large language model that supports the Model Context Protocol. It has two parts: a Blender add-on that runs a socket server inside Blender to receive and execute commands, and a Python MCP server that exposes those commands to AI clients such as Claude Desktop, Claude Code, Codex, Cursor and VS Code.
- Is mcp-for-blender free to use? What license does it have?
- mcp-for-blender is open source under the MIT license. Check the license terms before using it in commercial products.
- Is mcp-for-blender still maintained?
- Actively maintained. The last commit to mcp-for-blender was on Oct 18, 2026.
- What can I use mcp-for-blender for?
- Prototyping a 3D scene by describing it in natural language Populating a scene with HDRIs, textures and low-poly assets from free libraries Automating repetitive Blender modeling or layout tasks with an AI assistant Generating a starting 3D model from a text prompt or reference image
- How do I install mcp-for-blender?
- Run the one-line installer (macOS/Linux) Or, if uv is already installed, run setup: uvx mcp-for-blender setup Fully quit and reopen your AI app, then open Blender; the addon starts its server automatically
- Where can I get mcp-for-blender?
- The source code and installation instructions are on GitHub: https://github.com/ahujasid/mcp-for-blender
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Local-first AI memory system that stores conversation history verbatim and retrieves it with semantic search, exposed as an MCP server and CLI.
A minimal Python library for building AI agents that write and run their actions as Python code.
An open-source AI coding agent that runs in your terminal, with companion desktop, web, editor, and chat interfaces.
Agent-native memory infrastructure that turns LLM agent conversations and execution into persistent, structured memory for Python and TypeScript apps.
An agent framework that connects digital humans (2.5D, 3D, mobile, PC, web) and LLMs (OpenAI-compatible, DeepSeek) to business systems.