service-layer-extractor
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely informational, providing best-practice templates for software architecture. All code examples are standard patterns for business logic separation.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided source code for refactoring. While this represents a data ingestion surface, the skill lacks execution capabilities (eval/exec) or network access to non-whitelisted domains that would allow for exploitation via data poisoning. The instructions focus on structural transformation and do not introduce vulnerabilities.
Audit Metadata