refactoring-safely
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses standard CLI tools (cargo, clippy, git) to manage the refactoring process. These commands are recommended to be run via a dedicated test-runner agent, providing an execution boundary.
- [PROMPT_INJECTION]: No evidence of system prompt overrides, jailbreak attempts, or safety filter bypasses was found in the skill's instructions.
- [DATA_EXFILTRATION]: The skill does not access sensitive local files or make unauthorized external network requests. Its operations are focused on local file modifications and version control.
- [EXTERNAL_DOWNLOADS]: There are no references to external URLs for downloading or executing remote code or scripts.
Audit Metadata