evaluation-domains

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists entirely of documentation and configuration profiles that define how an AI agent should evaluate external information using the CRAAP (Currency, Relevance, Authority, Accuracy, Purpose) framework.
  • [SAFE]: The skill incorporates security-conscious design by providing a 'personal-digest' stub that explicitly forbids the storage of personal or sensitive data within the repository.
  • [SAFE]: It defines an 'Evidence-audit' procedure that requires the agent to verify every citation against actual source files in a fresh context, which serves as a robust defense against indirect prompt injection or misinformation in external documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 09:40 PM
Security Audit — agent-trust-hub — evaluation-domains