i18n-localization
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides documentation and code examples for internationalization. It leverages standard APIs and libraries without introducing malicious logic.
- [SAFE]: Code snippets include safety-relevant configurations like limiting supported languages, which helps prevent path traversal vulnerabilities when loading translation files.
- [INDIRECT_PROMPT_INJECTION]: The skill documents the processing of external translation files, which presents a surface for indirect prompt injection.
- Ingestion points: External JSON files (e.g., locales/en.json) loaded via dynamic imports or filesystem backends in SKILL.md.
- Boundary markers: None identified for distinguishing translation content from agent instructions.
- Capability inventory: The skill uses these inputs for text display and formatting; no dangerous tool execution is linked to this data in the provided snippets.
- Sanitization: Standard library interpolation is shown, which focuses on UI safety (XSS prevention) rather than instruction filtering.
Audit Metadata