deslop
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is guided by a 'fidelity contract' that mandates strict adherence to source facts and prohibits the creation of fabricated information or experiences.
- [SAFE]: Privacy safeguards are built into the instructions, specifically barring the agent from accessing or storing unauthorized voice samples from the user's environment.
- [SAFE]: During file modification tasks, the skill contains explicit directives to protect machine-readable sections like YAML frontmatter and code blocks from being altered.
- [SAFE]: The skill includes a clear refusal policy regarding requests for AI detector evasion, ensuring the tool is used for editorial quality rather than deceptive purposes.
Audit Metadata