anti-ai-copy
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill establishes clear UX writing principles derived from established design systems like Apple HIG and Google Material Design to improve user interface copy.
- [SAFE]: The instructions include a specific safety section ("改代码库时不能动的字") that prohibits the AI from modifying sensitive components like program logic strings, i18n keys, and internal agent instructions, which protects the application's functional integrity.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and modify external source code files. While this represents an attack surface, the risk is mitigated by explicit instructions for the AI to provide a comparison table for human oversight before applying any changes.
Audit Metadata