cro-optimization
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is entirely documentation-based, providing methodologies and templates for growth testing without any executable scripts, external downloads, or network activity.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external analytics and funnel data provided by the user. While this creates an ingestion surface, the impact is negligible as the skill lacks executable capabilities or tool access to exploit.
- Ingestion points: Funnel data, segmentation data, and analytics mentioned in the Required Inputs section of SKILL.md.
- Boundary markers: No specific delimiters or safety instructions are defined for the ingested data.
- Capability inventory: No tool usage or script execution patterns identified across the files.
- Sanitization: No data validation or sanitization logic is present.
Audit Metadata