Thanks for your interest in contributing! This skill brings Google's TurboQuant compression algorithm to Claude Code and compatible AI coding agents.
- Use the GitHub Issues tab
- Include your environment (Python version, OS, Claude Code version)
- For accuracy issues, include the dimension, bit-width, and a minimal reproduction
- Fork the repository
- Create a feature branch:
git checkout -b feat/your-feature - Make your changes
- Run the self-test:
python scripts/turboquant.py - Run the eval suite:
pytest evals/ -v - Commit with a descriptive message:
git commit -m "feat: add batch compression" - Push and open a Pull Request
- Performance improvements: GPU acceleration (CuPy/PyTorch), vectorized batch operations
- New quantization modes: 1-bit extreme compression, mixed-precision per-layer
- Integration examples: HuggingFace transformers, vLLM, TGI
- Documentation: Better explanations, more examples, benchmarks on real models
- Tests: Edge cases, numerical stability, large-dimension behavior
- Follow PEP 8 with a 100-character line limit
- Use type annotations for all public functions
- Docstrings for all public classes and methods
- NumPy-style docstrings preferred
This project follows the Anthropic Agent Skills specification.
When modifying SKILL.md:
- Keep the YAML frontmatter description specific and "pushy" (helps Claude trigger correctly)
- Keep the markdown body under 500 lines
- Move detailed content to
references/and link from SKILL.md
By contributing, you agree that your contributions will be licensed under the MIT License.