light-prototype
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a guide for generating prototype code and does not contain any executable scripts or malicious logic itself.
- [COMMAND_EXECUTION]: The instructions reference standard development tools and commands (pnpm, python, bun, etc.) as methods to run the generated prototypes. This is consistent with the skill's primary purpose as an engineering tool.
- [DATA_EXPOSURE]: The skill explicitly instructs the agent to avoid persistence by default and recommends using scratch databases or local files only when strictly necessary, which aligns with data minimization principles.
- [INDIRECT_PROMPT_INJECTION]: While the UI prototyping section suggests using URL search parameters (
?variant=) to toggle between different views, the implementation details describe static component switching which does not present a significant surface for indirect injection attacks.
Audit Metadata