principle-explain-the-number

Pass

Audited by Gen Agent Trust Hub on Nov 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill contains only markdown-based instructional content. It provides a methodological framework for performance testing and evaluation verification without any implementation of executable code. Comprehensive analysis of all 11 threat categories, including prompt injection, remote code execution, and data exfiltration, confirms the absence of any malicious patterns or technical security risks. The skill uses a platform configuration parameter 'disable-model-invocation: true', which further restricts its execution surface and ensures it serves only as static context.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 17, 2026, 09:36 AM