cs-refactor-ff
Pass
Audited by Gen Agent Trust Hub on Apr 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions define a process that limits operations to local file modifications and standard verification commands (grep, test runners) within a small, user-confirmed scope.
- [SAFE]: No network access, credential usage, or external code downloads are requested or performed by the instructions.
- [SAFE]: The skill requires a characterization test or existing test suite to verify changes, which is a defensive coding practice that mitigates risks of accidental behavior changes.
- [SAFE]: The skill processes untrusted user input and code files, creating an ingestion surface. It manages this through a mandatory 'alignment' phase where the agent must declare its intent and receive user confirmation before executing changes. Ingestion points: User prompts and local codebase. Boundary markers: Explicit confirmation step required. Capabilities: Local file system writes and test command execution. Sanitization: Verification via test suites.
Audit Metadata