deepchem
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were detected in the skill instructions or associated scripts.
- [EXTERNAL_DOWNLOADS]: The skill fetches pre-trained molecular models (such as ChemBERTa and MolFormer) from Hugging Face's model hub. Hugging Face is a well-known and established service for machine learning assets.
- [COMMAND_EXECUTION]: Includes Python scripts for automated model training and evaluation. These scripts use standard system libraries for argument parsing and do not contain arbitrary shell execution or unsafe input handling.
Audit Metadata