help-code
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided code and project briefs, which could contain instructions designed to manipulate the agent.
- Ingestion points: User-implemented code and task briefs referenced in SKILL.md.
- Boundary markers: The skill does not define specific delimiters to isolate user data from agent instructions.
- Capability inventory: The skill utilizes a subagent (luna) with reasoning capabilities to analyze ingested content.
- Sanitization: There is no specification for sanitizing or filtering the user-provided implementation before analysis.
Audit Metadata