reviewing-security

Pass

Audited by Gen Agent Trust Hub on Nov 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a well-structured security auditing tool. It implements the 'Goldfish' pattern for independent review, ensuring the reviewing agent starts with a fresh context and lacks write/edit permissions. It provides clear instructions to treat PR text and diffs as data rather than instructions, mitigating potential prompt injection from audited code. All external references are local to the skill's directory or target official developer tools (git, gh). The skill enforces least privilege by explicitly restricting subagents from using file-editing tools and limiting shell access to read-only commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 21, 2026, 01:16 AM