i18n-copy-and-layout
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill acts as a comprehensive documentation resource for developers working on multi-language support. It includes guidance on using standard browser APIs like
Intland ICU message formats. All external links point to reputable sources including W3C, Unicode, and MDN. - [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to analyze and review user-supplied UI code and translated text for localization errors. This functionality represents an ingestion point for potentially untrusted data. However, the skill does not include any capabilities for command execution, network access, or file modification, rendering the risk of code or instruction injection negligible.
- Ingestion points: UI source code (e.g., JSX, HTML) and user-facing text strings provided in prompts or files as seen in
evals/evals.jsonand thePR-worthiness gateinSKILL.md. - Boundary markers: None explicitly defined in the skill instructions.
- Capability inventory: No dangerous capabilities such as
exec(),eval(), network requests, or shell command execution are present in any of the skill's scripts or instructions. - Sanitization: Not applicable as the skill does not execute or store the processed data.
Audit Metadata