audit-agent-skill
Pass
Audited by Gen Agent Trust Hub on Aug 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill content is purely documentation and instructions for performing security audits. It contains no executable code, remote downloads, or malicious patterns.
- [PROMPT_INJECTION]: The skill is designed to process untrusted third-party data as its primary function. Ingestion points: External repository files,
SKILL.mddocuments, and referenced scripts. Boundary markers: The instructions explicitly mandate treating all skill content as untrusted and warn against exposing credentials. Capability inventory: Actions are restricted to static, non-executing checks like YAML parsing, link validation, and syntax checking. Sanitization: The methodology requires manual or agent-led inspection without execution by default to mitigate risks.
Audit Metadata