umap-learn
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides documentation and examples for using the legitimate umap-learn library for dimensionality reduction.
- [EXTERNAL_DOWNLOADS]: The skill installs the umap-learn package and its dependencies (scikit-learn, hdbscan, tensorflow) from the official Python Package Index (PyPI) using uv pip.
- [DYNAMIC_EXECUTION]: The skill utilizes the numba library for Just-In-Time (JIT) compilation of custom distance metrics, which is a standard performance optimization in the UMAP library.
- [DYNAMIC_EXECUTION]: The skill supports Parametric UMAP, which involves training neural network models using TensorFlow/Keras to learn embedding functions.
- [SAFE]: The skill includes security best practices by warning users about Python module shadowing, specifically advising against naming local files with package names like umap.py.
Audit Metadata