error-handling
Pass
Audited by Gen Agent Trust Hub on Mar 7, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes external codebase content during its audit phase, which presents a surface for indirect prompt injection.
- Ingestion points: codebase files (SKILL.md)
- Boundary markers: none
- Capability inventory: code generation and implementation in TypeScript and Python
- Sanitization: none
Audit Metadata