clean-code
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: No security issues detected. The skill consists exclusively of reference documentation and code examples provided in Markdown format for educational and advisory purposes.
- [NO_CODE]: The skill does not include any executable scripts, binaries, tool configurations, or shell commands. It relies entirely on providing information to the AI agent.
- [PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection as it is designed to analyze and review untrusted user-provided source code. While the interpolation of external content creates this surface, the risk is negligible as the skill defines no execution capabilities (network, file-system, or tool access). (1) Ingestion points: User-provided source code for review and refactoring (SKILL.md). (2) Boundary markers: Absent; no specific delimiters are defined for the code blocks being processed. (3) Capability inventory: None; no tools or scripts are included in the skill. (4) Sanitization: Absent; input is intended for structural and stylistic review by the agent.
Audit Metadata