deepchem
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data access attempts were detected in the provided skill file. The content focuses exclusively on the legitimate use of the DeepChem library for scientific research.
- [EXTERNAL_DOWNLOADS]: The skill recommends installing the 'deepchem' library and its associated dependencies (e.g., [torch], [all]) from official package registries using standard tools. This is expected behavior for a technical library skill.
- [EXTERNAL_DOWNLOADS]: The skill references pre-trained models from Hugging Face (e.g., 'seyonec/ChemBERTa-zinc-base-v1'), which is an established and well-known service for machine learning models.
- [COMMAND_EXECUTION]: The provided Python code snippets demonstrate standard, safe usage of the DeepChem API for loading data, featurizing molecules, and training machine learning models.
Audit Metadata