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
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:17 PM