refactor-assistant
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- PROMPT_INJECTION (LOW): Indirect Prompt Injection Surface. The skill is designed to ingest and process external source code which constitutes untrusted data. A malicious actor could embed instructions within comments or strings of the code being refactored to influence the agent's behavior.
- Ingestion points: Target code files and directories (e.g., UserService.js, src/).
- Boundary markers: None specified in the prompt to distinguish between code and data.
- Capability inventory: File system modification (implementation of refactoring changes).
- Sanitization: No sanitization or validation of the input code is described.
- NO_CODE (SAFE): No executable scripts (.py, .js, .sh) or configuration files were provided with this skill, reducing the direct execution risk.
Audit Metadata