Is Text-to-CAD Open Source? License, Code Access, and Usage Guide

Yes, text-to-CAD is fully open-source under the MIT License, allowing free use, modification, and distribution of the entire codebase.

The text-to-CAD project by earthtojake is publicly hosted on GitHub and released under one of the most permissive open-source licenses available. This means developers can inspect every line of code, fork the repository for custom workflows, and contribute improvements via pull requests without restrictive commercial limitations.

MIT License Terms and Permissions

The project is governed by the MIT License, as explicitly stated in the repository's root LICENSE file. This license, copyrighted 2026 by earthtojake, grants broad permissions while requiring only the preservation of the original copyright notice and license text in derivative works.

Key permissions granted by the MIT License include:

• Commercial use – You can integrate text-to-CAD into proprietary products
• Modification – Freedom to fork and alter the codebase for specific needs
• Distribution – Ability to share the original or modified versions publicly or privately
• Private use – No obligation to disclose changes when used internally

Repository Structure and Key Files

Understanding the layout of earthtojake/text-to-cad helps navigate the open-source resources effectively. The repository is organized into documentation, core packages, skill specifications, and validation scripts.

Core Documentation Files

The primary entry points for understanding the project are:

• [README.md](https://github.com/earthtojake/text-to-cad/blob/main/README.md) – Front-page documentation covering installation procedures, skill overviews, and visual benchmarks
• LICENSE – Complete MIT license text confirming open-source status
• [AGENTS.md](https://github.com/earthtojake/text-to-cad/blob/main/AGENTS.md) – Repository-level rules, layout standards, and contribution workflows for automated agents and human contributors

Source Code Packages

The reusable Python utilities reside in the packages/cadpy/ directory. This package contains helper functions for generating STEP files, GLB models, and topology artifacts that power the text-to-CAD conversion pipeline.

Skills and Specifications

The repository implements modular skills defined in markdown specifications:

• [skills/cad/SKILL.md](https://github.com/earthtojake/text-to-cad/blob/main/skills/cad/SKILL.md) – Core CAD generation skill specification
• [skills/urdf/SKILL.md](https://github.com/earthtojake/text-to-cad/blob/main/skills/urdf/SKILL.md) – URDF (Unified Robot Description Format) generation skill for robotic applications

How to Install and Use the Open Source Code

Because text-to-CAD is open source, you can install and run the code locally without proprietary restrictions.

Installing via Skills CLI

The fastest way to install the text-to-CAD skill set from the public repository is using the Skills CLI:

npx skills install earthtojake/text-to-cad

This command pulls the latest version of the CAD generation capabilities directly from the open-source repository.

Programmatic CAD Generation with Python

The cadpy package allows direct programmatic access to CAD operations. Located in packages/cadpy/, this library enables scripting of geometric primitives and export to industry-standard formats:

from cadpy import CadModel

model = CadModel()
model.box(width=100, depth=60, height=20)      # 100 × 60 × 20 mm block

model.export_step("block.step")

This example creates a simple rectangular block and exports it as a STEP file, demonstrating the library's integration with mechanical CAD workflows.

Running Repository Benchmarks

The benchmarks/ directory contains example CAD models and accompanying GIFs that showcase the library's capabilities. To view a specific benchmark:

git lfs pull --include="benchmarks/**"   # fetch heavy assets (optional)

open benchmarks/10-planetary-gear-stage.md

These benchmarks serve as both documentation and validation tests for the open-source algorithms.

Contributing to the Project

Open-source contributions follow the workflow defined in [AGENTS.md](https://github.com/earthtojake/text-to-cad/blob/main/AGENTS.md). The project uses [scripts/test/test.sh](https://github.com/earthtojake/text-to-cad/blob/main/scripts/test/test.sh) as the CI test harness to validate code changes, ensuring that contributions meet quality standards before merge.

Summary

• text-to-CAD is released under the MIT License, making it fully open source and commercially permissive
• The complete codebase is available in the earthtojake/text-to-cad GitHub repository with no restricted files
• Core functionality resides in the packages/cadpy/ Python package, supporting STEP and GLB export
• Skills are modularly defined in skills/cad/SKILL.md and skills/urdf/SKILL.md for CAD and robotics applications
• Installation is supported via npx skills install or direct repository cloning
• Contributions are managed through pull requests with validation via scripts/test/test.sh

Frequently Asked Questions

What open-source license does text-to-CAD use?

The project uses the MIT License, as documented in the root LICENSE file. This permissive license allows free use, modification, and distribution provided the original copyright notice and permission notice are included in all copies.

Can I commercially redistribute modified versions of text-to-CAD?

Yes. The MIT License explicitly permits commercial use and redistribution of modified versions. You can integrate the cadpy package or the skill definitions into proprietary products, provided you retain the original copyright notice and license text from the LICENSE file.

Where are the contribution guidelines for the text-to-CAD repository?

Contribution workflows and repository layout rules are defined in [AGENTS.md](https://github.com/earthtojake/text-to-cad/blob/main/AGENTS.md). This file specifies standards for both human contributors and automated agents, including how to structure pull requests and validate changes using the scripts/test/test.sh test harness.

How do I access the Python CAD utilities in the open-source repository?

The Python utilities are located in the packages/cadpy/ directory. You can import the CadModel class from this package to programmatically generate geometry and export to STEP format, as implemented in the open-source codebase and demonstrated in the repository examples.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}