llm-council

v1.0.1

Orchestrate multiple open-weight LLMs via Fireworks AI to deliberate on queries using Karpathy's LLM Council concept. Models respond individually, rank each other's responses, then a Chairman synthesizes the best answer. Powered entirely by fast, affordable open-weight models on Fireworks.

Claude Code1 skill

by DAIR.AI600updated 2 weeks ago

Source

$ git clone https://github.com/dair-ai/dair-academy-plugins

Clone the source, then follow the repository's marketplace instructions for your runtime. The plugin root is plugins/llm-council/ inside the repository.

Layout

plugins/llm-council/

├── .claude-plugin/plugin.json
└── skills/llm-council/SKILL.md

Skills1

llm-councilskills/llm-council/SKILL.md

Orchestrate multiple open-weight LLMs via Fireworks AI to deliberate on queries. Models respond individually, rank each other's responses, then a Chairman synthesizes the final answer. Use this skill when the user wants multiple AI perspectives, consensus-building, or the "LLM Council" approach inspired by Karpathy. Powered by fast, affordable open-weight models on Fireworks.

Manifests1

plugins/llm-council/.claude-plugin/plugin.json

{
  "name": "llm-council",
  "description": "Orchestrate multiple open-weight LLMs via Fireworks AI to deliberate on queries using Karpathy's LLM Council concept. Models respond individually, rank each other's responses, then a Chairman synthesizes the best answer. Powered entirely by fast, affordable open-weight models on Fireworks.",
  "version": "1.0.1",
  "author": {
    "name": "DAIR.AI"
  }
}