agent-safety-guard

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The file SKILL.md contains conceptual educational material and markdown/code templates for protecting AI agents against prompt injection and data leakage. The mention of prompt injection phrases, such as 'Ignore previous instructions', is entirely contextual within a defensive example allowlist and testing checklist. No executable dangerous behavior, exfiltration vectors, or actual malicious instructions are present.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 05:12 AM
Security Audit — agent-trust-hub — agent-safety-guard