skills/simota/agent-skills/polyglot/Gen Agent Trust Hub

polyglot

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for installing well-known development libraries such as i18next, react-intl, and vue-i18n using npm. It also describes the integration of the lingo.dev CLI for translation management.
  • [EXTERNAL_DOWNLOADS]: The documentation references the i18n-actions/ai-i18n GitHub Action, which is used to automate machine translation processes within a CI/CD environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill architecture handles user-facing strings that could serve as a vector for indirect prompt injection when processed by AI translation models. Ingestion points: Hardcoded strings, mobile resource files (iOS/Android), and translator context comments are the primary data entry points. Boundary markers: The skill advocates for ICU MessageFormat syntax, which uses curly brace delimiters for variable interpolation and plural logic. Capability inventory: The skill facilitates the extraction of code strings and their transmission to external translation services, including AI-driven pipelines. Sanitization: To address potential risks, the skill explicitly requires human review for all machine-generated translations, particularly for legal or safety-critical content.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 01:06 PM