Mojulo

Mojulo is a 3D compiler for coding agents: an MCP server that runs wherever your agent runs, on your machine or in the throwaway Linux box it gives itself, where the agent you already run (Claude Code, Codex, any MCP host, or a chat app that can open a Linux box: Claude, ChatGPT with Codex, Grok, Meta Muse) builds objects, walkable worlds and games by conversation, and what gets stored is source, not a mesh. Every artifact is a small deterministic recipe: a few hundred bytes of JSON, or an OpenSCAD program, that a kernel compiles back to the same geometry on every read, byte for byte, and that emits to Godot, Blender, Unity, Unreal, glTF, OpenUSD, or print-ready STL / 3MF at true scale. Small enough to carry home from a box and re-mint on your own disk. A compiler, not a generator: you edit and diff the recipe like code, and renders are disposable. No API key, no account, no telemetry. Your agent does the thinking; mojulo holds the state and does the geometry.

Quickstart

npx mojulo init

Needs Node 22.12+ and an MCP-capable coding agent (Claude Code or Codex; Claude Desktop works too). init finds the hosts on your machine, asks once per host, and opens the dashboard at http://localhost:3001. Everything lands in ~/.mojulo/. Then open a fresh agent session and ask: "what is this?"

The first install is the big one: npx pulls a ~27 MB tarball whose native dependencies land at about 590 MB on disk (2.0.6 measured 885 MB before the local search model and its runtime became the opt-in mojulo install recall; semantic_search ranks lexically without them). Most of the rest is puppeteer-core (local headless Chrome for stills) and better-sqlite3. Nothing in the list reaches the network on its own; the per-dependency sheet is in docs/tech-requirements.md. Verified on macOS (Apple Silicon) and on native Windows under Claude Code (init and a first render). On Linux x64 the test suite runs in CI, and the agent-box path below has been run to a mint and an export from the Claude app and web, ChatGPT work mode with Codex, Grok chat and Meta Muse.

# Claude Code (--scope user makes mojulo available in every project):
claude mcp add --scope user mojulo -- npx -y mojulo

# Codex: add to ~/.codex/config.toml
[mcp_servers.mojulo]
command = "npx"
args = ["-y", "mojulo"]

// Claude Desktop: add under "mcpServers" in claude_desktop_config.json, then restart.
// If it fails with "spawn npx ENOENT", replace "npx" with the absolute path from `which npx`.
"mojulo": { "command": "npx", "args": ["-y", "mojulo"] }

Where it runs, three things you can add

Mojulo runs wherever your agent runs, and a recipe minted in one place re-mints byte for byte in the other. Two shapes, one install.

In your agent's box, nothing on your machine. One sentence to the agent installs mojulo in the throwaway Linux box it gives itself:

• The Claude app (macOS, Windows, web, iOS); Grok chat; Meta Muse (iOS, web, macOS app, one session across all three): Open a Linux box and install the mojulo npm package in it.
• ChatGPT (Codex included): In work mode, open a Linux box and install the mojulo npm package in it.

Each of those has been run this way, from the sentence to a mint and an export handed back. You get the same recipes and the same exports, as files: the export result names this host's door, an artifact page, a PR or a file card. The box has no dashboard you can reach, a scene-to-PNG bake needs a browser it may not be allowed to fetch, and it is gone when the session ends, so ask for the bundle (one zip: world.html, mesh, print STL for literal-scale objects, recipe.json, README) and keep the recipe. A box with no MCP client never sends initialize, so tell the agent to run npx mojulo orient first: it prints what an MCP client is handed at connect, translated to the shell, and points at the routing index. Claude's box built a 47-part phone at true scale this way and handed back the glTF; Grok chat's sandbox minted a city from the shell. Blender installs in those boxes too.

On your machine: macOS, Windows, Linux. npx mojulo init wires mojulo into the agents it finds, opens the dashboard at localhost:3001, keeps everything under ~/.mojulo/, and probes your PATH for the optional local workers (Blender, a slicer, OpenSCAD, the game engines). This is the whole loop, and where a recipe from a box comes home to.

Who has run it where. persistent (your machine; ~/.mojulo/ stays) · ephemeral (the web agent's throwaway Linux box; keep the recipe). Blank means not verified yet, not "does not work".

agent	macOS	Windows	Web Agent Linux Box (Headless)
Claude Code
Claude app (macOS, Windows, web, iOS)
ChatGPT (work mode; Codex)	Codex
Hermes Agent
Grok (Build; chat)	Build
Meta Muse (iOS, web, macOS app; one session across them)

Three things are opt-in, and the choice is the same in both places:

mojulo install with no argument prints which of the three are present.

Six things to say to it

Each one is a sentence to your agent, the tool it reaches for, and the recipe that gets stored. Every example below runs keyless and offline.

1. "Make me a coffee mug, 90 mm tall"

The agent calls mint_solid { kind: 'workbench' }. This is the whole stored recipe: a lathe for the body, a sweep for the handle, real centimetres.

{ "units": "cm",
  "lathes": [ { "axisFrom": { "x": 0, "y": 0, "z": 0 }, "axisTo": { "x": 0, "y": 0, "z": 9 },
                "profile": [ { "t": 0, "radius": 3.6 }, { "t": 0.15, "radius": 4 }, { "t": 1, "radius": 4 } ],
                "tint": "#b8342c", "material": "satin" } ],
  "sweeps": [ { "path": [ [3.8, 0, 2.2], [6.2, 0, 3.2], [6.6, 0, 5.2], [5.4, 0, 7.0], [3.8, 0, 7.4] ],
                "radius": 0.6, "tint": "#b8342c", "material": "satin" } ] }

It is served shaded and turnable at /sketches/<ref>, and the same ref downloads as model.stl or model.3mf at 90 mm tall, z-up, slicer-ready. "Make the handle thicker" is update_sketch setting one radius; nothing is re-minted. measure_solid reads wall thickness, overhangs and closure off the recipe before you print, and if a slicer is installed the print gate stamps layers, time and filament beside the file. A bore through the body is a cuts line naming the body and the sweep that bores it; exact: true on that line composes it with Manifold instead of the sampled grid, so the lip is a true circle and a 40 mm disc measures 40, in the viewer, the .glb and the print.

2. "Write me a Raspberry Pi 4 case tray in OpenSCAD"

The agent calls mint_solid { kind: 'scad' } and the program is the recipe, stored verbatim: