ia-refine-prompt
Pass
Audited by Gen Agent Trust Hub on Jul 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a structured methodology for prompt engineering without using any dangerous operations or accessing sensitive data.
- [SAFE]: Instructions explicitly prohibit the refinement of prompts intended for harmful or illegal tasks, adhering to safety best practices.
- [SAFE]: File persistence functionality is secured by requiring explicit user confirmation before writing to the local
.ai/PROMPT.mdfile. - [SAFE]: Technical documentation in SPEC.md correctly identifies and prohibits the storage of secrets, credentials, or machine-specific paths within the skill components.
Audit Metadata