karpathy-guardrails

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill contains no executable code, scripts, or external dependencies. It serves as a set of meta-instructions for software development tasks.
  • [SAFE]: The instructions focus on establishing software engineering best practices and do not attempt to bypass agent safety filters or exfiltrate data.
  • [SAFE]: The use of project-specific file paths for logging (e.g., .cavekit/history/backprop-log.md) is within the expected scope of a development tool and does not involve sensitive system configuration files or credentials.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 09:28 PM