clean-architecture-refactor
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external codebases during its analysis and refactoring phases, which could contain malicious instructions designed to influence agent behavior.
- Ingestion points: Reads files and project structures through Explore subagents to map existing architecture (SKILL.md).
- Boundary markers: Explicitly requires an approval gate in Phase 3, where the user must review and approve the migration plan before any code is modified.
- Capability inventory: Performs file system modifications (moving/editing code), updates namespaces/references, and executes build and test commands (SKILL.md).
- Sanitization: There is no explicit sanitization of input code, relying instead on the human-in-the-loop approval process to verify intended changes.
Audit Metadata