skills.rest
abzhaw
abzhawCOMMUNITY
Explore all skills by this author →
GitHub

juliaz-agent-architect

Guide architectural decisions for the Juliaz multi-agent system using hub-and-spoke, ReAct loop, and gateway patterns.

Updated Feb 21, 2026
Search Tags:#architecture#patterns#multi-agent#gateway#hub-and-spoke#react-loop#juliaz-agents
One-click install
npx skills add https://github.com/abzhaw/juliaz_agents --skill juliaz-agent-architect
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: juliaz-agent-architect
Source: https://github.com/abzhaw/juliaz_agents/tree/main/.claude/skills/juliaz-agent-architect
Command: npx skills add https://github.com/abzhaw/juliaz_agents --skill juliaz-agent-architect

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Designing and evolving a complex multi-agent system like Juliaz_agents can be error-prone without a clear architectural playbook. This skill offers proven patterns and anti-pattern guidance to align decisions with robust, scalable structures.

Core Features & Use Cases

  • ReAct Loop (Orchestrator + Frontend) guidance for structured reasoning and tool use.
  • Hub-and-Spoke message routing through Bridge, OpenClaw gateway integration, and centralized orchestration.
  • Anti-Patterns to Avoid with concrete recommendations to prevent common pitfalls.
  • Decision Framework for New Agent vs. New Tool, including criteria for ownership, state, and scheduling.
  • Clear architectural rules and file mappings to keep team alignment and implementation consistent.

Quick Start

Explain how to apply hub-and-spoke, ReAct loop, and gateway patterns to restructure Juliaz_agent architecture.

Frequently Asked Questions about juliaz-agent-architect

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I design a robust multi-agent architecture using hub-and-spoke patterns?▼

To design a robust multi-agent architecture, apply the hub-and-spoke pattern for centralized message routing through a Bridge and gateway, ensuring structured orchestration and scalable system behavior. This prevents ad-hoc agent communication.

What is the ReAct loop pattern for multi-agent systems?▼

The ReAct loop pattern integrates an orchestrator and frontend to govern structured reasoning and tool use. It guides multi-agent systems through iterative cycles of reasoning, acting, and observing to execute complex tasks.

When should I add a new agent versus a new tool in a multi-agent architecture?▼

Decide whether to add a new agent or a new tool by evaluating criteria for ownership, state, and scheduling. Create a new agent when distinct state or scheduling autonomy is required; otherwise, register a new tool.

What are common anti-patterns to avoid when building multi-agent message flows?▼

Common multi-agent anti-patterns to avoid include bypassing the gateway for direct agent communication and breaking the ReAct loop. Following explicit architectural rules and file mappings prevents these structural pitfalls.

Does the Juliaz multi-agent system require a gateway for OpenClaw integration?▼

Yes, the Juliaz multi-agent system requires the OpenClaw gateway for hub-and-spoke message routing. The gateway centralizes orchestration and bridges frontend interactions with backend agent execution.

Related Skills

Explore More Software Engineering Skills→
HechaoYannetHechaoYannet

create-architecture

Generate a master architecture document from GDDs, ADRs, and governance constraints.

Community
Advanced
alsk1992alsk1992
650

routing

Route messages to specialized agents and enforce tool access policies.

Community
Advanced
chemistrywow31chemistrywow31
54

Team Topology Analysis

Analyzes agent graph topology for safe handoffs and parallel work in Codex multi-agent systems.

Community
Intermediate
shreed27shreed27

botchan

Post messages to onchain feeds and send direct messages between agents on Base.

Community
Intermediate