human-gate-designer
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill documents an architectural pattern for 'Feedback Injection' that is susceptible to indirect prompt injection.
- Ingestion points: External data enters the agent context via the 'human feedback' text input during the modification phase (SKILL.md).
- Boundary markers: The template uses basic double quotes as delimiters, which are easily bypassed by adversarial input.
- Capability inventory: According to the frontmatter, the skill environment allows
Read,Write, andEdittools, which could be abused if an injected instruction is executed (SKILL.md). - Sanitization: There is no evidence of input escaping or instructions to ignore embedded commands within the feedback string.
- [SAFE]: No evidence of hardcoded credentials, unauthorized network communication, or persistent mechanisms was found. The skill primarily provides instructional content and design patterns.
Audit Metadata