deepchem
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate documentation and scripts for the DeepChem library. No malicious code, obfuscation, or persistence mechanisms were found. The metadata accurately reflects the functionality.
- [EXTERNAL_DOWNLOADS]: The skill facilitates the download of pretrained models from well-known Hugging Face repositories (e.g.,
seyonec/ChemBERTa-zinc-base-v1) and MoleculeNet benchmark datasets. These are reputable sources and the downloads are standard for the intended ML workflow. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface through file loaders (
CSVLoader,SDFLoader,FASTALoader) in the included scripts. While these ingest untrusted external data, the formats (SMILES, FASTA) are processed by scientific parsers and do not typically contain natural language instructions that could influence the agent's behavior. No explicit boundary markers or sanitization for prompt injection are present, but the capabilities are restricted to molecular data processing.
Audit Metadata