compound-component
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional, providing a senior architect persona and templates for React component refactoring. It does not request or use any system tools, network access, or file system operations.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes user-provided source code as an ingestion point, it lacks any high-risk capabilities (like shell execution or network exfiltration) that would make indirect injection exploitable. The instructions focus on static code transformation and UI patterns.
- Ingestion points: User-provided component code in Step 1 (SKILL.md).
- Boundary markers: Not explicitly defined.
- Capability inventory: Text-based code generation only.
- Sanitization: None required for the described UI patterns.
Audit Metadata