alterlab-gget

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes explicit examples and parameters that pass credentials verbatim (e.g., --api_key your_key_here, --email/--password for COSMIC, and gget.gpt(..., api_key="your_key_here")), which requires the LLM to handle and potentially emit secret values directly.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md and module references show it directly queries many public third-party data sources (e.g., gget.archs4 → ARCHS4 HTTP API, gget.cellxgene → CELLxGENE census, gget.enrichr → Enrichr, gget.cbio → cBioPortal, gget.cosmic → COSMIC, Ensembl/UniProt/NCBI REST APIs) and then programmatically reads and acts on those results (e.g., selecting hits, downloading PDBs, plotting, driving downstream analyses), so untrusted public content can materially influence tool behavior and thus enable indirect prompt injection.