code-review-refactor

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill mentions companion resources from a well-known service (Vercel Labs) for repository maintenance and multi-agent installation patterns.
  • [COMMAND_EXECUTION]: Utilizes standard Android development tools including Gradle for testing and linting (./gradlew test, ./gradlew lint), as well as a local Node.js validation script (node scripts/validate-skills.mjs). These are appropriate for the skill's intended purpose.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted user-supplied code and diffs while maintaining command execution capabilities.
  • Ingestion points: Processes diffs, files, and module trees provided by the user (found in SKILL.md).
  • Boundary markers: The instructions lack explicit delimiters or warnings to ignore instructions embedded within the code being reviewed.
  • Capability inventory: Executes shell commands via Gradle and Node.js (found in SKILL.md).
  • Sanitization: No specific sanitization or filtering for external code content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 12:30 PM
Security Audit — agent-trust-hub — code-review-refactor