lathe
Pass
Audited by Gen Agent Trust Hub on Jul 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, data exfiltration risks, or unauthorized execution methods were detected in the skill instructions or templates.- [SAFE]: The skill provides structured templates that follow industry best practices for agent safety, including the use of XML delimiters to prevent prompt injection and the requirement for independent verification of agent actions.- [SAFE]: All external references are to trusted organizations and well-known research within the AI community, such as Anthropic, Google, and OpenAI.
Audit Metadata