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
Risk Level
SAFE
Analyzed
Sep 17, 2026, 08:36 PM
Security Audit — agent-trust-hub — i18n-localization