agent-council

Collect and synthesize responses from multiple CLI-based AI agents.

8|Updated Jul 26, 2026

Search Tags:#workflow #multi-agent #orchestration #synthesis #decision-making #ai-collaboration

Ready to run

npx skills add https://github.com/joonlab/joonlab-claudecode-setting-for-share --skill agent-council-joonlab

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agent-council
Source: https://github.com/joonlab/joonlab-claudecode-setting-for-share/tree/main/claude/skills/agent-council
Command: npx skills add https://github.com/joonlab/joonlab-claudecode-setting-for-share --skill agent-council-joonlab

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires node, yaml, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill solves the problem of single-model bias by allowing you to query multiple AI agents simultaneously and synthesize their diverse perspectives into a single, well-rounded answer.

Core Features & Use Cases

Multi-Agent Synthesis: Orchestrates parallel execution of prompts across different AI models or CLI tools.
Configurable Council: Easily define your own panel of experts (e.g., Claude, Codex, Gemini) via a simple YAML configuration.
Use Case: When performing a complex architectural review or technical decision, use this Skill to gather opinions from multiple AI agents to identify blind spots and reach a more robust conclusion.

Quick Start

Ask the agent council to evaluate the pros and cons of using a microservices architecture for this project.

Frequently Asked Questions about agent-council

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

FAQPage Schema

How do I synthesize multiple AI agent perspectives for complex decision-making?▼

To synthesize multiple AI agent perspectives, this Skill dispatches your query to configured CLI-based AI members in parallel and collects their responses to generate a comprehensive, multi-perspective answer for collaborative decision-making.

How do I configure a multi-agent council using a YAML file?▼

You configure a multi-agent council by defining a panel of CLI-based AI experts, such as Claude or Gemini, within a YAML configuration file, which the Skill uses to orchestrate parallel agent tasks and synthesize responses.

Do I need Node.js to orchestrate parallel AI agent prompts?▼

Yes, you need Node.js installed to execute this Skill, as it requires the Node.js runtime to dispatch prompts to configured CLI-based AI members and facilitate multi-agent orchestration.

What is the best way to avoid single-model bias during architectural reviews?▼

The best way to avoid single-model bias during architectural reviews is to query multiple AI agents simultaneously, allowing you to identify blind spots and synthesize diverse perspectives into a more robust conclusion.

Can I use this multi-agent synthesis workflow for technical architectural reviews?▼

Yes, you can use this multi-agent synthesis workflow for technical architectural reviews, as it gathers opinions from multiple AI agents to evaluate pros and cons and reach a well-rounded conclusion.