prompt-optimizer

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a reference repository for prompt engineering methodologies, including 57 individual framework files (e.g., RACEF, CRISPE, BAB, and Tree of Thought) used to guide the generation of high-quality AI instructions.
  • [SAFE]: No malicious patterns related to prompt injection, data exfiltration, or persistence mechanisms were detected in the instructions or framework documentation.
  • [SAFE]: The skill lacks network or shell capabilities; its primary function is to read local markdown files and generate text-based suggestions for the user.
  • [SAFE]: External URLs within the framework files point to legitimate educational resources on the juuzt.ai domain and are not used for automated downloads or remote code execution.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 02:59 PM
Security Audit — agent-trust-hub — prompt-optimizer