ai-auditing-code

Pass

Audited by Gen Agent Trust Hub on May 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a structured framework for auditing AI implementation code using the DSPy library. It checks for signature design, module composition, data pipeline integrity, metric design, optimizer usage, production readiness, and common anti-patterns.
  • [SAFE]: All external references, such as the recommended installation command for the ai-do skill, are directed toward the author's own resources (lebsral), which represents standard vendor functionality.
  • [SAFE]: The skill uses clear instructional language and does not attempt to bypass safety filters or exfiltrate data. It strictly operates on user-provided code files after explicit consent.
  • [SAFE]: No obfuscation, hidden commands, or suspicious data handling patterns were found in the instructions, examples, or reference materials.
Audit Metadata
Risk Level
SAFE
Analyzed
May 8, 2026, 10:53 AM