clean-arch
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill is designed to perform architectural reviews of code repositories using a sub-agent for exploration and provides findings based on an internal reference document.
- [SAFE]: Indirect Prompt Injection Surface: The skill ingests untrusted repository data via the Explore sub-agent in Phase 2. Ingestion points: Repository scan (SKILL.md). Boundary markers: None. Capability inventory: Repository mapping and findings report generation (SKILL.md). Sanitization: None. This is a common pattern for analysis tools and does not represent a malicious intent.
Audit Metadata