cove-prompt-design

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill is entirely composed of documentation and instructional templates for improving LLM factual accuracy through prompting.
  • [SAFE]: No executable scripts, command-line operations, or system-level modifications are present.
  • [SAFE]: The skill does not reference external URLs, download remote content, or include third-party dependencies.
  • [SAFE]: Analysis of the instructions revealed no attempts at prompt injection, obfuscation, or data exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 08:40 AM