pytdc-therapeutics-data-commons
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses the
PyTDClibrary, which automatically downloads AI-ready datasets (ranging from 10-500 MB) from the official Therapeutics Data Commons servers upon first access. This is the primary intended function of the library for drug discovery research. - [DYNAMIC_EXECUTION]: The skill makes use of molecular oracles and format conversion utilities (
MolConvert). These tools use pre-trained machine learning models or rule-based logic for chemical property prediction and representation changes, which do not involve the execution of arbitrary or untrusted system code. - [DATA_EXFILTRATION]: There are no indicators of sensitive data access or unauthorized network transmissions. The skill focuses on reading chemical data and protein sequences for local model training and evaluation.
- [PROMPT_INJECTION]: The instructions are strictly focused on providing technical documentation and code examples for the PyTDC library. No attempts to override agent constraints or bypass safety filters are present.
Audit Metadata