ncbi-datasets-skill

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [External Service Interaction]: The skill fetches genomic and taxonomic metadata from the NCBI Datasets API (api.ncbi.nlm.nih.gov). This is a well-known scientific resource, and the script hardcodes the base URL to ensure requests are directed to the legitimate service.
  • [Local File Writing]: The script includes functionality to save raw API responses to the local file system. While the default behavior uses /tmp, it allows specifying a raw_output_path. This capability for file creation and modification should be monitored in secure environments.
  • [Indirect Prompt Injection Surface]: As the skill ingests data from an external scientific database, it is subject to indirect prompt injection considerations. The script implements basic mitigation via string truncation (_compact function) and structured data parsing, which reduces the risk of the agent misinterpreting external data as instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 12:21 PM
Security Audit — agent-trust-hub — ncbi-datasets-skill