tool-design-output
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains instructional content and metadata defining design principles for tool results. No executable code, scripts, or dangerous commands are present.
- [SAFE]: Example data, such as the Base64-encoded string used for a pagination cursor example, is benign and decodes to a simple JSON object (offset: 50) with no hidden instructions or URLs.
- [SAFE]: The skill configuration in agents/openai.yaml and its descriptive frontmatter align with its educational purpose and the author's platform (Runtype) without requesting excessive permissions or performing unauthorized operations.
Audit Metadata