Our take
Activeregular commits and releases- Activity
- 88
- Community
- 44
- Issues
- 75
- Pull requests
- 70
+5,198 stars in the last 30 days
- Releases more often than 92% of the projects we track
This is a broad, actively maintained agent-to-CAD workflow with useful export and fabrication tooling, but its young age and heavy contributor concentration make project continuity worth weighing.
Good fit if
- You already work with a supported coding agent and want it to generate build123d-based CAD scripts and common manufacturing exports.
- Your workflow benefits from combining CAD creation with drawing, DFM, printability or robotics skills.
- You are comfortable with the MIT-licensed project and its local installation approach.
Look elsewhere if
- You need a conventional standalone CAD application rather than an agent-driven Python-script workflow.
- Your team cannot accept a project where most contributions come from one person, despite its frequent releases and active issue handling.
- Your required agent or manufacturing workflow is not among the supported integrations and skills described by the project.
All health signals
| Last commit | 2026-11-02 (today) |
|---|---|
| Commits, last 90 days | 100+ |
| Releases, last 12 months | 93 (latest v0.10.0, 2026-11-02) |
| Contributors | 22 (top contributor: 80% of commits) |
| Issues closed, last 90 days | 14+ (typically closed in 1 day) |
| Pull requests merged, last 90 days | 28 (typically merged in 0 days) |
| Project age | 7 months |
Checked on 2026-11-03 with the GitHub API.
Project history
Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.
Stars
26,672Forks
2,542Change since Nov 17, 2026: +160 stars·+14 forks
Overview
text-to-cad gives AI agents local workflows for designing physical products as 3D models. Models are written as Python scripts with build123d, built on Open CASCADE, then compiled by text-to-cad's server into editable XBF documents and formats such as STEP, GLB, STL and 3MF. A viewer follows model rebuilds as an agent edits the design.
The project is aimed at people using coding agents for mechanical design and fabrication, rather than at replacing a general-purpose CAD application. It supports agents including Claude Code, Codex, Cursor, Gemini and Grok, and extends the workflow to manufacturing checks, drawings, robot-description files and fabrication services.
Key Features
- Generate and edit CAD models from plain-language or image requests through agent skills.
- Build Python scripts with build123d and export STEP, GLB, STL and 3MF files.
- Keep model parts, names, colours, materials and joints in an XBF document.
- Review models in an agent app or browser-based CAD viewer, with updates following rebuilds.
- Create dimensioned engineering drawings and 2D DXF files.
- Check models for printability and review designs for sheet metal, CNC machining or injection molding.
- Support adjacent robotics workflows with URDF, SRDF and SDF skills, plus G-code slicing and printer workflows.
Use Cases
- A product designer can ask a coding agent to draft a part or assembly, then inspect and revise the model without manually rebuilding each change.
- A mechanical engineer can export STEP files, generate a dimensioned drawing and check a design against manufacturing constraints.
- A robotics developer can create CAD geometry and related URDF, SRDF or SDF files as part of an agent-assisted workflow.
- A 3D-printing user can assess mesh printability, slice with OrcaSlicer presets and send prints to supported Bambu Lab workflows.
What you need
Detected in the repository
- Node.js (from package.json)
- A test suite and automated checks on GitHub Actions
License in plain words
MITpermissive
- Commercial use: yes
- Modify and redistribute: yes
- You must keep: the copyright and license notice
- Share your changes: no
A summary, not legal advice: the LICENSE file is what applies.
Getting Started
On macOS or Linux, install with:
curl -fsSL https://texttocad.dev/install.sh | sh
For Windows, use the PowerShell installer shown in the README. See the installation guide for manual setup and updates.
Considerations
The project is still young, even though it has an active commit and release cadence. Its contributor base is larger than a solo-maintained project, but most code contributions come from one person, which creates a continuity risk. The README says usage statistics and crash reports are enabled by default; they use a random ID and exclude files, paths and prompts, and the project links to instructions for opting out. The MIT license permits broad use, subject to its license terms.
Found this useful?
Share it with someone who would like text-to-cad.