deliberation

Facilitate multi-agent architectural deliberation among heterogeneous LLMs via GitHub Discussions.

3|Updated Feb 15, 2026

Search Tags:#design #architecture #multi-agent #deliberation #LLM debate #GitHub Discussions

Ready to run

npx skills add https://github.com/pvliesdonk/agents.md --skill deliberation

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: deliberation
Source: https://github.com/pvliesdonk/agents.md/tree/main/skills/deliberation
Command: npx skills add https://github.com/pvliesdonk/agents.md --skill deliberation

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses complex architectural and design problems by facilitating a structured debate among multiple AI models (Claude, Gemini, GPT) to arrive at well-reasoned decisions.

Core Features & Use Cases

Heterogeneous Model Debate: Leverages the strengths of different LLMs to explore diverse perspectives on design challenges.
GitHub Discussions Integration: Uses GitHub Discussions as a persistent, auditable platform for deliberation, promoting transparency and collaboration.
Use Case: When deciding on a new microservice architecture, use this Skill to have Claude, Gemini, and GPT debate the pros and cons of different approaches, considering factors like scalability, maintainability, and cost.

Quick Start

Initiate a deliberation on the architectural choice for the new user authentication module.

Frequently Asked Questions about deliberation

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

FAQPage Schema

How do I use multi-agent LLM debate for complex architectural design decisions?▼

Multi-agent LLM debate for architectural design uses heterogeneous models to explore diverse perspectives. It enables AI models to challenge assumptions and provide evidence-based recommendations by debating trade-offs on complex design problems.

How does LLM debate work with GitHub Discussions?▼

LLM debate via GitHub Discussions uses the platform as a persistent, auditable space for multi-agent deliberation. It requires GitHub API integration to post discussion threads where heterogeneous AI models debate architectural design choices transparently.

Do I need API keys for multiple LLMs to run an architectural deliberation?▼

Yes, running an architectural deliberation requires model APIs for LLM interaction with heterogeneous models. You also need GitHub API integration to facilitate the multi-agent discussion posting and enable AI models to debate design trade-offs.

What is the best way to evaluate microservice architecture trade-offs using AI?▼

Evaluating microservice architecture trade-offs using AI is best achieved through heterogeneous model debate. By having different LLMs challenge assumptions and debate factors like scalability and cost, you receive well-reasoned, evidence-based design recommendations.

Can I use multi-model debate for software design problems outside of microservices?▼

Yes, you can use multi-model debate for various complex software design problems outside of microservices. The deliberation mechanism addresses any architectural challenge by facilitating structured debate among heterogeneous LLMs to explore diverse perspectives and trade-offs.