eval-driven-development
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of markdown documentation and instructions without any executable scripts, tool calls, or external network activities.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential indirect prompt injection surface when using LLMs as judges for candidate outputs. \n
- Ingestion points: LLM candidate outputs being evaluated by a judge model (SKILL.md). \n
- Boundary markers: Recommends delimited/quoted blocks and explicit 'data, not instructions' framing. \n
- Capability inventory: No tool use or execution capabilities provided in this skill. \n
- Sanitization: Recommends structural isolation of untrusted content.
Audit Metadata