molfeat

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the molfeat library and its optional extras (e.g., molfeat[all], molfeat[dgl], molfeat[transformer]) directly from PyPI. It also identifies the map4 featurizer as an external dependency hosted on the official reymond-group GitHub repository.
  • [COMMAND_EXECUTION]: The instructions utilize uv pip for package management and standard shell utilities like grep for navigating reference documentation. These are routine development tasks consistent with the skill's purpose as a cheminformatics setup tool.
  • [DATA_EXPOSURE]: The skill includes explicit security best practices in its documentation, specifically advising against the use of pickle for caching due to insecure deserialization risks. It correctly suggests using NumPy's np.savez as a secure alternative for storing numerical embeddings.
  • [SAFE]: The referenced domains (datamol.io, valencelabs.com, pypi.org) and GitHub organizations (datamol-io, reymond-group) are established and reputable sources within the scientific and machine learning communities.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 06:23 AM