skills/jimmc414/kosmos/bioservices/Gen Agent Trust Hub

bioservices

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • SAFE (SAFE): No malicious patterns or unauthorized behaviors were detected. The skill relies on the established 'bioservices' Python package.
  • PROMPT_INJECTION (SAFE): While the skill possesses an indirect prompt injection surface through data ingestion from external bioinformatics APIs, the risk is negligible and inherent to the tool's function.
  • Ingestion points: Data retrieved via service wrappers like UniProt, KEGG, and QuickGO in SKILL.md.
  • Boundary markers: None present in the snippets.
  • Capability inventory: Network requests (HTTP GET/POST) and local script execution (via the scripts/ directory).
  • Sanitization: Not explicitly shown, as the skill provides a direct interface to raw scientific data.
  • DATA_EXFILTRATION (SAFE): Network operations are directed solely toward well-known bioinformatics web services (e.g., uniprot.org, kegg.jp) for data retrieval. These operations are essential to the skill's primary purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:18 PM