committee

Name: committee
Availability: InStock
Author: crispy13

Coordinate four AI models to research, debate, and produce a consensus report.

Updated Feb 27, 2026

Search Tags:#consensus #evidence #risk #deliberation #design-review #multimodel #committee

Ready to run

npx skills add https://github.com/Crispy13/copilot-directions --skill committee-crispy13

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: committee
Source: https://github.com/Crispy13/copilot-directions/tree/main/.github/skills/committee
Command: npx skills add https://github.com/Crispy13/copilot-directions --skill committee-crispy13

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

General committee coordinates four AI models to independently research, debate, and converge on a consensus report, enabling robust handling of ambiguous, high-stakes questions.

Core Features & Use Cases

Orchestrates four model subsystems (Opus, Sonnet, Gemini, GPT) to gather diverse perspectives.
Structured rounds of discussion to surface tradeoffs and reduce bias.
Produces a final consensus report with traceable reasoning, including evidence and risks.
Applicable to design reviews, RFCs, root-cause analyses, or tool/vendor comparisons requiring multiple viewpoints.

Quick Start

Invoke a committee run on your topic by loading Opus, Sonnet, Gemini, and GPT, and return a consolidated consensus report.

Frequently Asked Questions about committee

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

FAQPage Schema

How do I coordinate multiple AI models for design review and consensus?▼

You coordinate multiple AI models for design review by loading Opus, Sonnet, Gemini, and GPT as a committee to independently research, debate converging tradeoffs, and produce a structured consensus report with traceable reasoning.

What is multimodel deliberation for high-stakes decision making?▼

Multimodel deliberation is a process where four AI models independently research and debate ambiguous, high-stakes topics to surface tradeoffs, reduce bias, and converge on a final consensus document with provenance, evidence, and risks.

How do I run a root-cause analysis using diverse AI reasoning perspectives?▼

Run a root-cause analysis by invoking a four-model committee on your topic, which triggers parallel drafting across Opus, Sonnet, Gemini, and GPT, followed by chief consolidation into a structured final output.

Can I use this committee approach for RFC and tool comparison evaluations?▼

Yes, the committee approach suits RFC and tool comparison evaluations by orchestrating four model subsystems to gather diverse perspectives, surface tradeoffs through structured discussion rounds, and produce a traceable consensus report.

What are the limitations of using a four-model committee for consensus?▼

A limitation of the four-model committee is the requirement to load four specific subagents, meaning you must have access to Opus, Sonnet, Gemini, and GPT to execute the phased workflow and produce the final consolidated output.

When do I need a multimodel committee for consensus reports?▼

You need a multimodel committee for consensus reports when handling ambiguous or high-stakes topics such as design decisions, root-cause analysis, RFCs, and tool comparisons where diverse reasoning matters to reduce bias.