skills/i18n-agent/i18nstack/i18n-wrap/Gen Agent Trust Hub

i18n-wrap

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill triggers the execution of local shell commands for project validation, including pnpm build, pnpm lint, ng extract-i18n, and flutter gen-l10n. While these are standard developer tasks, they involve executing tools defined within the project environment.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from source code files and locale files during its scanning and wrapping phases (Phases 2 and 3). There are no boundary markers or sanitization procedures described to prevent malicious instructions embedded in the analyzed code from overriding the agent's behavior.
  • Ingestion points: Source files (JSX, TSX, Python, Ruby, etc.) and locale files (JSON, YAML, PO, etc.) as described in Phase 3 and Phase 6.
  • Boundary markers: None identified; the skill directly processes strings and attributes from source code.
  • Capability inventory: File read/write access and shell command execution for linting and building.
  • Sanitization: No evidence of input sanitization or validation of the content being wrapped.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 08:00 AM
Security Audit — agent-trust-hub — i18n-wrap