text-to-cad: what it is, what problem it solves & why it's gaining traction

What it solves

It provides a specialized set of tools (skills) that allow AI agents to handle the entire hardware design lifecycle—from generating 3D CAD models and robot description files to sourcing parts and preparing files for 3D printing or CNC cutting.

How it works

The project acts as a library of "skills" that can be installed into AI agents (such as Claude Code or Codex). These skills enable the agent to execute specific workflows, such as creating CAD models from plain-language or image requests, writing URDF/SDF robot files, finding off-the-shelf parts, and slicing meshes into G-code for 3D printers.

Who it’s for

It is designed for developers and engineers building AI agents for robotics, hardware design, and fabrication.

Highlights

• CAD Generation: Creates and edits models from text or images, exporting to STEP, STL, 3MF, and GLB.
• Robot Descriptions: Generates URDF, SRDF (for MoveIt2), and SDF files for simulation and structure.
• Fabrication Pipeline: Includes tools for DXF drawing creation, G-code slicing, and integration with Bambu Lab printers and SendCutSend.
• Part Sourcing: A dedicated skill for finding off-the-shelf STEP parts like motors and bearings.
• Visualization: Includes a CAD Viewer for browser-based previews of CAD, G-code, and robot files.