Council

Coordinate structured multi-agent debates with 3-round and 1-round workflows.

1|Updated Jan 24, 2026
One-click install
npx skills add https://github.com/verrio1/vaughn-pai --skill council-verrio1
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Council
Source: https://github.com/verrio1/vaughn-pai/tree/main/skills/Council
Command: npx skills add https://github.com/verrio1/vaughn-pai --skill council-verrio1

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables teams and systems to surface high-quality decisions by orchestrating structured, multi-agent debates. By formalizing roles, prompts, and interaction rounds, it reveals diverse viewpoints and surfaces actionable insights through transcripts.

Core Features & Use Cases

  • Multi-agent council with roles (Architect, Designer, Engineer, Researcher) and optional members.
  • Full debate workflow (3 rounds) plus a quick single-round check for fast validation.
  • Output includes a detailed transcript and a synthesis with recommended actions.
  • Customizable council composition and prompts to fit topics like architecture, UX, or data strategy. Use Case: Before a major architectural decision, run a DEBATE to surface concerns, trade-offs, and consensus.

Quick Start

Initiate a topic-based 3-round DEBATE workflow using the Council skill. Example: 'Council: Should we adopt WebSockets or SSE for real-time updates?' Return the complete 3-round transcript and final synthesis.

Frequently Asked Questions about Council

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

FAQPage Schema
How do I run a multi-agent debate for architectural decision-making?

To run a multi-agent debate, initiate a topic-based 3-round workflow using the Council skill. It coordinates agents across defined roles to surface trade-offs and returns a complete transcript with a final synthesis of recommended actions.

What is a multi-agent council debate and how does it work?

A multi-agent council debate is a structured workflow where agents assigned roles like Architect or Engineer discuss a topic over three rounds. It formalizes prompts and interactions to reveal diverse viewpoints and generate actionable insights.

Can I use a multi-agent debate workflow for UX and data strategy discussions?

Yes, you can use the debate workflow for UX, data strategy, and policy discussions. The council composition and prompts are customizable to fit specific topics, ensuring relevant viewpoints are surfaced during the decision-making process.

Is there a quick way to validate decisions without running a full 3-round debate?

Yes, there is a quick single-round check available for fast validation. This option provides a faster workflow to surface immediate concerns when you do not need the comprehensive output of a full 3-round debate transcript.

What format does the multi-agent debate output use for decision synthesis?

The debate output includes a detailed transcript of the interaction rounds and a final synthesis. This synthesis consolidates the viewpoints of all agents into a set of recommended actions for your decision-making context.