New:Socket for Asana Is Now Available.Learn more
Sign In

apolo-cad

Package Overview
Dependencies
Maintainers
1
Versions
1
Alerts
File Explorer

Advanced tools

Socket logo

Install Socket

Detect and block malicious and high-risk dependencies

Install

apolo-cad

CAD paramétrico 3D agente-nativo para maquinaria industrial, conducible por un agente de IA vía MCP

pipPyPI
Version
0.1.0
Weekly downloads
23
-39.47%
Maintainers
1
Weekly downloads
 
Created

Genix Apolo CAD

An agent-native 3D parametric CAD for industrial machinery. Built to be driven by an AI agent (Claude Opus or others) over MCP — or by hand in your browser.

A belt conveyor modeled in Apolo

Apolo is a headless parametric CAD for designing real machines. Its edge isn't the kernel (it uses OpenCascade, like FreeCAD) but its agent-native architecture: every operation is a command against an API, the whole document is an editable command log, and the same JSON Schemas that generate the UI also generate the agent's tools.

The upshot: an AI agent can design complete machines end to end —not just autocomplete— and verify them: detect interferences, simulate gravity, look at a render (vision) and emit fabricable shop drawings.

Install

pip install apolo-cad

Optional extras: apolo-cad[physics] (gravity/drop tests, MuJoCo), apolo-cad[fea] (linear static FEA), apolo-cad[sketch] (PlaneGCS solver where no wheel is published — needs a C++ toolchain; otherwise the sketcher falls back to the SciPy engine automatically).

Run

apolo --open          # API + web UI on http://127.0.0.1:8000

Use it from an AI agent (MCP)

Apolo exposes 79 MCP tools, so any MCP-compatible client (Claude Code, Claude Desktop, …) can design machines. Keep apolo running, then register the stdio server:

{
  "mcpServers": {
    "apolo-cad": {
      "command": "apolo-mcp",
      "env": { "APOLO_URL": "http://127.0.0.1:8000" }
    }
  }
}

Then just ask:

"Design a 4 m × 600 mm belt conveyor for 1–15 kg parcels, with a hollow-shaft gearmotor and gravity-type take-up tensioning. Check there are no interferences and show me a render."

The agent models (run_batch — atomic batches, one undo step), perceives (render_view returns an image → vision), validates (check_interference, engineering_check, gravity_test) and documents (drawing_set, assembly_manual → shop drawings, cut lists, BOMs, assembly manuals). Everything lives in the command log: editable, undoable, reproducible.

What it does

  • Modeling — primitives, fillet/chamfer/shell/drill, patterns, mirror, revolve, sweep/loft, sheet metal with flat-pattern DXF/SVG, constrained 2D sketching, STEP import.
  • Assembly & kinematics — persistent face mates, joints, rail/N-DOF constraints, motion studies with collision scanning.
  • Library & BOM — a 231-reference catalog from real standards (ISO/ASTM/DIN/EN) + machine super-commands (belt conveyor, weldment, frame, sheet metal, robot arm).
  • Validation — engineering rules, OCCT interference booleans, and gravity-based assembly validation (simulates what falls).
  • Pro manufacturing drawings — HLR → SVG/DXF/PDF, sections, auto hole dimensioning, drawing sets, exploded views, assembly manuals, GD&T and ISO 2553/2768/1302 shop annotation.
  • FEA (linear static) — per part, and bonded multi-material assembly (whole welded frame, safety factor per piece).
  • Engineering deliverables — calculation report (A4 PDF citing CEMA/ISO/DIN/EN/AISC per check), costed BOM + quotation, tolerance stack-up, installation sheet, and a green/amber/red delivery check before you ship a design.

Data location

Installed from PyPI, your projects live in ~/.apolo/ (override with APOLO_HOME, or point APOLO_DB at a specific SQLite file). Running from a source checkout, they stay in the repo's data/ — same as before.

  • Source & docs: https://github.com/mariorojasmz/apolo-cad
  • Issues: https://github.com/mariorojasmz/apolo-cad/issues

MIT © 2026 Mario Rojas. Built on OpenCascade (LGPL), build123d, FastAPI, three.js and MuJoCo.

Keywords

cad

FAQs

Related posts