skills.rest
RooseveltAdvisors
RooseveltAdvisorsCOMMUNITY
Explore all skills by this author →
GitHub

Council

Orchestrate multi-agent debates with structured rounds and specialized roles.

1|1|Updated Mar 1, 2026
Search Tags:#decision making#collaboration#multi-agent#perspectives#synthesis#debate
One-click install
npx skills add https://github.com/RooseveltAdvisors/claude-agent-stack --skill council-rooseveltadvisors
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Council
Source: https://github.com/RooseveltAdvisors/claude-agent-stack/tree/main/skills/Thinking/Council
Command: npx skills add https://github.com/RooseveltAdvisors/claude-agent-stack --skill council-rooseveltadvisors

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and scripts (resource) components.

What problem does it solve?

This Skill facilitates collaborative decision-making by simulating a debate among specialized AI agents, ensuring diverse perspectives are considered before a final path is chosen.

Core Features & Use Cases

  • Structured Debate: Engages multiple AI agents (Architect, Designer, Engineer, Researcher) in a 3-round debate format.
  • Collaborative Adversarial Analysis: Agents respond to and challenge each other's points, fostering intellectual friction for deeper insights.
  • Quick Consensus Checks: Offers a faster, single-round option for rapid perspective gathering.
  • Use Case: When deciding on a new software architecture, use the "Council: Evaluate architecture options for Project X" command to get input from all relevant AI perspectives, leading to a more robust and well-vetted decision.

Quick Start

Run a full council debate on the topic of implementing a new caching strategy.

Frequently Asked Questions about Council

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

FAQPage Schema
How does multi-agent debate improve software architecture decision making?▼

Multi-agent debate improves architecture decisions by simulating specialized AI roles like Architect and Engineer. These agents challenge each other's points in structured rounds, creating intellectual friction that synthesizes diverse perspectives into a robust, well-vetted conclusion.

How do I run a collaborative analysis of architecture options using AI perspectives?▼

You can run a collaborative analysis by initiating a full 3-round council debate. This structured format forces specialized agents to respond to and challenge each other, ensuring comprehensive coverage before synthesizing a final decision on your architecture options.

Can I get rapid feedback on a decision without running a full multi-agent debate?▼

Yes, you can use the quick consensus check feature to get rapid feedback. This performs a faster, single-round debate that rapidly gathers diverse AI perspectives without the time investment of a full 3-round deliberation.

What specialized AI roles are used to evaluate software engineering decisions?▼

The debate utilizes specialized agent roles including Architect, Designer, Engineer, and Researcher. These distinct perspectives ensure comprehensive coverage of your software engineering topics by evaluating the decision from multiple professional angles.

When should I avoid using a multi-agent debate for decision making?▼

You should avoid multi-agent debates for simple or highly urgent decisions that do not benefit from adversarial analysis. If a topic does not require synthesizing diverse perspectives or deep deliberation, the debate format adds unnecessary processing overhead.

Related Skills

Explore More Software Engineering Skills→
leonlettoleonletto
28

thrum-prime

Loads a complete session briefing for AI agents in a workflow.

Community
Intermediate
rysweetrysweet

debate-workflow

Facilitate structured multi-perspective debates for architectural decisions.

Community
Advanced
ShinyGuaShinyGua

mm-discussion-moderator

Synthesize analyst memos and cross-critique files into thesis_map.json and debate_summary.md.

Community
Advanced
CAPHTECHCAPHTECH
1

lde

Design evaluation tests probing failure modes in LLM-based systems.

Official
Advanced