review-code

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed as a read-only analysis engine that performs structured audits of code diffs. It explicitly prohibits modification of repository files, mitigating risks related to unauthorized write access.
  • [SAFE]: The use of external tools such as WebFetch and WebSearch is strictly limited to researching API documentation and verifying contracts, which aligns with the skill's primary purpose.
  • [SAFE]: Thorough inspection of all 14 files identified no evidence of prompt injection, obfuscation, hardcoded credentials, or malicious command execution.
  • [SAFE]: While the skill ingests untrusted content (source code and external web documentation), the risk of indirect prompt injection is minimized by the skill's use of specific analysis roles (e.g., defect-finder), structured reporting requirements, and a lack of executable capabilities beyond research and diffing.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 09:47 PM
Security Audit — agent-trust-hub — review-code