refactor-assistant
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes external code files that may contain adversarial instructions disguised as comments or string literals.
- Ingestion points: The agent reads and analyzes source code files (e.g., UserService.js) as part of its primary function.
- Boundary markers: There are no explicit instructions to the agent to disregard natural language instructions found within the code comments or data.
- Capability inventory: The skill allows the agent to read files, propose structural changes, and incrementally write modifications to the file system.
- Sanitization: The skill does not implement validation or sanitization of the input code to prevent embedded instructions from influencing agent behavior.
Audit Metadata