code-review

Pass

Audited by Gen Agent Trust Hub on Jun 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill defines a clear and legitimate purpose for reviewing code changes without any suspicious commands or exfiltration patterns.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface by processing untrusted code data.
  • Ingestion points: The @$1 argument in SKILL.md receives external code diffs or file content.
  • Boundary markers: No explicit delimiters are used to separate the instructions from the untrusted code content.
  • Capability inventory: The skill is restricted to file system tools (Read, Glob, Grep) and lacks high-risk tools like code execution or network access.
  • Sanitization: The skill does not perform validation or sanitization on the ingested content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 14, 2026, 08:50 AM
Security Audit — agent-trust-hub — code-review