cook-backend
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: No executable code, scripts, or remote dependencies are included in the skill.
- [SAFE]: The skill promotes secure development practices by mandating security reviews for SQL injection, exposed secrets, and authentication checks during the [REVIEW] phase.
- [SAFE]: Data persistence is handled locally via markdown files in a specific 'memory-bank' directory, with no external network exfiltration paths or non-whitelisted domain requests.
- [SAFE]: The instructional content does not contain any prompt injection patterns, bypass attempts, or instructions to ignore safety guidelines.
- [SAFE]: The skill processes project requirements and documentation files to maintain session state. While this represents a surface for indirect prompt injection, it is the primary intended functionality for session continuity and does not possess capabilities for privilege escalation or unauthorized network access.
Audit Metadata