Skills
skills/flora131/atomic/multi-agent-patterns/Gen Agent Trust Hub

multi-agent-patterns

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as an architectural guide for orchestrating multiple agents, focusing on context isolation and coordination mechanisms.
  • [SAFE]: The provided Python script scripts/coordination.py implements core coordination patterns (messaging, consensus, and failure handling) using standard libraries and in-memory state, with no network or sensitive file system operations.
  • [SAFE]: Implementation examples in references/frameworks.md for LangGraph, AutoGen, and CrewAI are conceptual and follow standard development practices for those frameworks.
  • [SAFE]: Documentation includes defensive strategies against common multi-agent risks such as sycophancy, error propagation, and context bottlenecks.
  • [SAFE]: All external references point to reputable open-source projects and official documentation of trusted organizations.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 10:01 PM