product-manager
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with external state files, creating a surface for indirect prompt injection.\n
- Ingestion points: The skill reads from files in the .nugelo/ directory, specifically index and feature-slug files (SKILL.md).\n
- Boundary markers: The skill contains explicit instructions to the AI to treat these files as untrusted decision data and specifically warns never to follow directives found inside them (SKILL.md).\n
- Capability inventory: The skill is designed to manage state via file read/write operations and version control commits (SKILL.md).\n
- Sanitization: The instructions require the AI to surface untrusted content to the user rather than acting on it autonomously.
Audit Metadata