human-gate-designer
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill content is restricted to markdown documentation and Mermaid diagrams describing workflow design.
- [SAFE]: No scripts, binary files, or external software dependencies are included or referenced for installation.
- [SAFE]: The analysis found no evidence of credential usage, network operations, or persistence mechanisms.
- [PROMPT_INJECTION]: The skill describes a pattern for injecting human feedback into subsequent prompts, which is an indirect prompt injection surface.
- Ingestion points: 'Human feedback' field in the 'Feedback Injection' section of SKILL.md.
- Boundary markers: Uses simple quotation marks and brackets for template interpolation.
- Capability inventory: None (design documentation only).
- Sanitization: Not mentioned in the implementation guidelines.
Audit Metadata