source-handbook
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's functionality is limited to reading its own internal reference file using the
readFiletool to answer policy questions. No external network requests, sensitive file access, or remote code execution patterns are present. - [SAFE]: Analysis of the skill metadata and reference content found no evidence of prompt injection, obfuscation, or deceptive instructions. The stated purpose of the skill aligns with its implementation.
Audit Metadata