ln-61-skill-reviewer

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a structured methodology for auditing AI agent skills, emphasizing repository-specific conventions, standalone behavior, and structural integrity.
  • [SAFE]: The instructions mandate a strict read-only boundary, explicitly stating that the agent should not edit files, stage changes, or modify external state during the review process.
  • [SAFE]: It includes security-conscious instructions for handling evidence, requiring that commands and outputs be recorded without exposing sensitive information or secrets.
  • [SAFE]: The skill utilizes legitimate repository-defined validators and official documentation as information sources, which is consistent with its stated purpose of release readiness review.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 01:49 PM
Security Audit — agent-trust-hub — ln-61-skill-reviewer