agent-designer

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as an architectural design toolkit for multi-agent systems, providing conceptual frameworks for patterns, communication, and memory.
  • [SAFE]: Analysis of both README.md and SKILL.md reveals no evidence of prompt injection, data exfiltration, or unauthorized command execution.
  • [SAFE]: The skill actively encourages security-first design, specifically requiring human-in-the-loop checkpoints and the principle of least privilege for agent tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 08:06 PM
Security Audit — agent-trust-hub — agent-designer