systematic-testing
Pass
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious behaviors, obfuscation, or safety bypass attempts were detected within the skill content.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of external data (user code and logs) as its primary function for testing and debugging.
- Ingestion points: User-provided source code, error logs, and bug descriptions (SKILL.md).
- Boundary markers: None provided; the skill relies on the agent's base instruction set to maintain context isolation.
- Capability inventory: Recommends the use of standard development tools like
npm,git, andtailfor diagnostics (SKILL.md). - Sanitization: No explicit input sanitization or validation logic is defined within the markdown instructions.
Audit Metadata