roo-translation

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The workflow requires the execution of a local validation script node scripts/find-missing-translations.js to ensure all locales are updated.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes and translates strings from project files that could contain malicious instructions.
  • Ingestion points: Local JSON translation files in src/i18n/locales/ and webview-ui/src/i18n/locales/.
  • Boundary markers: No specific delimiters or instructions are provided to the agent to ignore embedded commands within the strings being translated.
  • Capability inventory: The skill allows file modification using apply_diff and command execution via Node.js.
  • Sanitization: No sanitization or content validation is described for the strings before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 06:18 PM