alterlab-deepchem
Warn
Audited by Snyk on Apr 12, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.70). The skill explicitly loads public MoleculeNet datasets (dc.molnet.load_* used in SKILL.md and scripts like scripts/graph_neural_network.py) and downloads pretrained models from third-party hubs (dc.models.HuggingFaceModel with 'seyonec/ChemBERTa-...' in scripts/transfer_learning.py and GroverModel), which are untrusted public sources and are ingested at runtime to drive model training/prediction and thus can materially influence agent behavior.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata