Skip to content
MCP server

Mojulo

MCP server that compiles small recipes into 3D objects, worlds and games for coding agents

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, macOS, Windows
Runs
Locally
Version
3.1.1
Listing updated
today

Similar MCP servers

About Mojulo

Mojulo is an open-source MCP server for design and media. It runs locally on your machine and is started by your AI client.

Mojulo stores source rather than meshes. Each artifact is a deterministic recipe, a few hundred bytes of JSON or an OpenSCAD program, which a kernel compiles to the same geometry on every read, byte for byte on the same platform. Recipes can be edited and diffed like code and exported to Godot, Blender, Unity, Unreal, glTF, OpenUSD, or print-ready STL and 3MF at true scale. The README calls it a compiler rather than a generator, with the agent supplying the thinking.

It runs locally through npx mojulo init, which detects Claude Code, Codex and Claude Desktop, opens a dashboard on localhost:3001 and keeps data under ~/.mojulo. It can also be installed inside the throwaway Linux box that agents such as Claude, ChatGPT with Codex, Grok, Meta Muse and Google AI Studio provide. There is no API key, account or external telemetry. It needs Node 22.14 or newer, and the first install is large, about 227 MB on disk.

Optional extras include a semantic-search model installed on request. The chatbot factory from 2.x was removed in 3.0.

Install

Run in a terminal

claude mcp add mojulo -- npx -y mojulo

Add to claude_desktop_config.json

{
  "mcpServers": {
    "mojulo": {
      "command": "npx",
      "args": [
        "-y",
        "mojulo"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "mojulo": {
      "command": "npx",
      "args": [
        "-y",
        "mojulo"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "mojulo": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "mojulo"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "mojulo": {
      "command": "npx",
      "args": [
        "-y",
        "mojulo"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "mojulo": {
      "command": "npx",
      "args": [
        "-y",
        "mojulo"
      ]
    }
  }
}

Using several servers? Build one config file with the MCP config generator.

What it does

Deterministic recipes
Small JSON or OpenSCAD sources compile to identical geometry.
Many export formats
Godot, Blender, Unity, Unreal, glTF, OpenUSD, STL and 3MF.
Runs in agent boxes
Install in a hosted Linux sandbox with no local setup.
Local dashboard
Browse results at localhost:3001.
No account
No API key or external telemetry.
Claude Code plugin
Install as a marketplace plugin pinned to a version.

Good for

  • Have an agent design a printable part at true scale
  • Generate a walkable world and export it to Godot
  • Keep 3D assets as editable recipes in Git
  • Produce print-ready STL files from a conversation

Pricing

Open source

Mojulo FAQ

What does it need?

Node 22.14 or newer and an MCP-capable coding agent; the first install is about 227 MB on disk.

Does it need an API key?

No. The README says no key, no account and no external telemetry.

Can I use it without a local install?

Yes, ask a web agent to install the npm package in its Linux box, though the box may be wiped when the session ends.

What formats can it export?

Godot, Blender, Unity, Unreal, glTF, OpenUSD, STL and 3MF.

Is Mojulo free?

Yes. Mojulo is open source, so you can use and self-host it for free.

How do I add Mojulo to Claude Code?

Run this command in your terminal: claude mcp add mojulo -- npx -y mojulo

What are the best alternatives to Mojulo?

The closest alternatives to Mojulo are HyperFrames by HeyGen, next-ai-draw-io, mcp-for-blender, ffmpeg-skill and cursor-talk-to-figma-mcp.

Where this comes from

  • Description: written from the project's README.
  • Install steps: from the official MCP Registry entry.

How AIPoweredBot checks servers