council

Query Claude Opus, Gemini, and GPT-5.4 in parallel and synthesize responses into a dated session file.

Updated Dec 17, 2025

Search Tags:#logging #strategy #consensus #parallel #council #decision-making #multimodel

Ready to run

npx skills add https://github.com/eprouveze/HealthPulse --skill council-eprouveze

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: council
Source: https://github.com/eprouveze/HealthPulse/tree/main/.agent/skills/council
Command: npx skills add https://github.com/eprouveze/HealthPulse --skill council-eprouveze

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Enables teams to gather diverse strategic advice by querying Claude Opus, Gemini, and GPT-5.4 in parallel and synthesizing a unified recommendation.

Core Features & Use Cases

Parallel tri-model inquiry: Send the same question to three advisors and collect their responses.
Persistent session logs: Save questions and outputs to docs/council-sessions for auditability.
Synthesized consensus: Lex aggregates and presents a unified conclusion with noted disagreements when relevant.
Use Case: Before high-stakes decisions, run a council to compare viewpoints and extract a robust recommendation.

Quick Start

Ask a question with /council <your question> to receive three model responses and a synthesized consensus stored in docs/council-sessions.

Frequently Asked Questions about council

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

FAQPage Schema

How do I get consensus from multiple LLMs for strategic decision-making?▼

To achieve multi-model consensus for strategic decision-making, you query Claude Opus, Gemini, and GPT-5.4 in parallel with the same question. The tool then synthesizes their independent responses into a single unified recommendation, highlighting any disagreements to provide a robust conclusion.

How do I log LLM advisory sessions for auditability?▼

To log LLM advisory sessions for auditability, the tool automatically creates a dated session directory under docs/council-sessions. It saves the original question and each model's reply as separate files, generating a concise synthesis file to ensure a persistent, auditable record of the parallel inquiry.

What is the best way to compare risk assessments from different AI models?▼

The best way to compare risk assessments from different AI models is using a parallel tri-model inquiry. By sending the same policy or risk question to Claude Opus, Gemini, and GPT-5.4 simultaneously, you can directly compare their distinct viewpoints and extract a synthesized consensus.

Can I use parallel AI model querying for product and design considerations?▼

Yes, you can use parallel AI model querying for product and design considerations. The orchestration scope explicitly covers strategic decision-making, risk assessment, and policy guidance across product, business, and design contexts, aggregating diverse advice into a single synthesized view.

Do I need to install external dependencies to run tri-LLM consensus?▼

No, you do not need to install external dependencies to run tri-LLM consensus. The Skill operates without any listed dependencies, requiring only a prompt to trigger the orchestration of Claude Opus, Gemini, and GPT-5.4, followed by the automatic generation of session logs.