molfeat

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFE
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill documents the use of MoleculeTransformer.from_state_yaml_file and ModelStore.load (using joblib). These functions involve deserializing configuration or model weights, which can execute arbitrary code if the input file is malicious. The skill provides clear guidance to only load trusted artifacts and highlights the risks of arbitrary third-party state.
  • [REMOTE_CODE_EXECUTION]: The documentation mentions the MolJEPATransformer adapter, which requires enabling the 'trust-remote-code' flag for specific Hugging Face models. This is a known security-sensitive feature of the transformers library used to execute model-specific code provided by authors.
  • [EXTERNAL_DOWNLOADS]: The skill performs legitimate network operations to fetch model metadata from https://fs.molfeat.datamol.io/artifacts/ and model weights from the Hugging Face Hub. These are well-known, domain-specific services essential for the skill's primary functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 09:46 PM