Products/CAD/Developer Tools/agentcad

agentcad

open-source CAD tool for AI agents

CAD/Developer Tools

Our Take

This is one of those ideas that makes you do a double-take: a CAD tool built for AI agents to use, not humans. The entire interface is designed for coding agents to manipulate and create 3D models, which sounds unhinged until you realize every founder is now building AI products that need to interact with physical space somehow. AgentCAD is free, open source, and honestly the kind of infrastructure play that either becomes essential in 18 months or disappears entirely — the vibes are very "early mover building the pipes before anyone knows they need plumbing."

An open-source MCP server and CLI that lets your coding agent design, render, and export real CAD output. Agents write Python code to create 3D CAD models with built-in primitives and get immediate feedback through renders and geometric metrics.

Key Features

MCP server integration for AI agents, CLI with no boilerplate required, Execute: run scripts, produce versioned STEP files + geometric metrics, Render: PNG views from any angle for visual verification, Export: STL, GLB, OBJ for 3D printing and web viewers, Validate: pre-execution checks catch errors in <100ms, Inspect: topology report for debugging geometry issues, Diff: compare versions to track design iteration, Pre-injected build123d primitives and show_object helper

Problem It Solves

AI coding agents typically write bad geometry on the first try. agentcad provides a tight feedback loop — run, render, inspect, fix — so agents converge on printable 3D geometry without human intervention babysitting them through every step.

Target Customer

Developers building with AI coding agents (Claude Code, Cursor, Windsurf) who need automated CAD design capabilities

Use Cases

3D modeling automation for AI agents, CAD design workflow automation, Rapid prototyping with AI assistance, 3D printing model preparation

Pricing Details

Open source, free, no signup required, runs locally

Free Tier

Full product is free and open source (Apache-2.0)

Differentiator

Built specifically for human–agent collaboration with tight feedback loops. Every run returns metrics and a render so the agent reads both and catches its own mistakes before they reach the human.

Why Now

AI coding agents are becoming more capable and need specialized CAD tools to automate 3D design workflows

Traction

Notable Metrics: 234 parts built in Wright Flyer case study

Key Facts

The people behind agentcad

James Dillard

profile

Builder

GitHub

Links

Website GitHub Source: product-hunt

Want products like this in your inbox every morning?