gastroenterology

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill consists of natural language instructions and editorial criteria for evaluating manuscript fit for the Gastroenterology journal. No scripts, binaries, or complex automation patterns are present.
  • [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to read local resource files (../../resources/source-basis.md and ../../resources/official-source-map.md) and perform web searches for the latest journal guidelines. This ingestion of external data is the intended function of the skill and does not introduce malicious capabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 11:33 AM
Security Audit — agent-trust-hub — gastroenterology