translate-multilingual
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses local Node.js scripts (e.g., prepare-translation.mjs and merge-translation.mjs) for task preparation and data merging. These scripts utilize Node.js built-in modules ('node:fs', 'node:path') and perform no network operations.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local project files for translation, creating a potential vector for indirect prompt injection. 1. Ingestion points: Locale JSON files in 'rosey/locales/' and Markdown/MDX files in 'src/content/'. 2. Boundary markers: The workflow uses structured JSON manifests to separate content from instructions. 3. Capability inventory: The skill has file-writing capabilities via the merge scripts. 4. Sanitization: The scripts include HTML validation logic to verify that translations maintain the original tag structure.
Audit Metadata