skills/kursku/skills/pubmed-database/Gen Agent Trust Hub

pubmed-database

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill connects to the National Center for Biotechnology Information (NCBI) servers at eutils.ncbi.nlm.nih.gov to retrieve scientific metadata and abstracts. This interaction is limited to a well-known, trusted government service for public research data.
  • [PROMPT_INJECTION]: The skill processes article abstracts from an external database, representing a standard architectural surface for indirect prompt injection. No malicious patterns or exploits were found.
  • Ingestion points: Data retrieved from ESearch and EFetch API endpoints.
  • Boundary markers: None used in the provided search result examples.
  • Capability inventory: Network operations restricted to HTTP GET requests using the requests library.
  • Sanitization: Not applicable as the data is used for informational research purposes without execution capabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 04:32 PM
Security Audit — agent-trust-hub — pubmed-database