LLM Council — A Claude Code Skill

Stop trusting Claude's first answer. Run any decision through 5 AI advisors who argue, peer-review each other anonymously, and hand you a verdict you can actually trust.

Based on Andrej Karpathy's LLM Council methodology, adapted to run entirely inside Claude Code using sub-agents with different thinking styles.

The Problem

Claude is incredibly agreeable. Ask it "should I launch this product?" and it'll find 5 reasons why you should. Ask "is this product a bad idea?" and it'll find 5 reasons why it is. Same product, different framing, opposite answers.

That's fine for writing emails. It's dangerous for making decisions.

How It Works

When you say "council this", the skill:

1. Scans your workspace for relevant context (CLAUDE.md, memory files, etc.)
2. Frames your question into a neutral prompt
3. Spawns 5 advisors in parallel, each with a different thinking style:
   • The Contrarian — hunts for what will fail
   • The First Principles Thinker — asks if you're solving the right problem
   • The Expansionist — looks for upside you're missing
   • The Outsider — responds with zero context (catches curse of knowledge)
   • The Executor — only cares what you do Monday morning
4. Anonymizes their responses and runs a peer review — advisors review each other without knowing who said what
5. Chairman synthesizes the verdict: where the council agrees, where it clashes, blind spots it caught, a clear recommendation, and one concrete next step
6. Generates a visual HTML report + full markdown transcript

All in one session. About 4 minutes.

Install

Option 1 — Git clone (recommended)

git clone https://github.com/tenfoldmarc/llm-council-skill ~/.claude/skills/llm-council

Then open Claude Code (claude in Terminal).

Option 2 — Manual

1. Create folder ~/.claude/skills/llm-council/
2. Drop SKILL.md inside it
3. Restart Claude Code

Use

Type any of these triggers followed by your question:

• council this
• run the council
• pressure-test this
• stress-test this
• war room this
• debate this

Example:

council this: I'm thinking of pivoting from a $297 course to a $97 live workshop for my audience of non-technical solopreneurs. Is that the right move?

Give it context. The richer the input, the sharper the output.

You'll get:

• A visual HTML report that opens automatically
• A full markdown transcript saved alongside it

When To Use It

Good council questions:

• "Should I launch a $97 workshop or a $497 course?"
• "Which of these 3 positioning angles is strongest?"
• "I'm thinking of pivoting from X to Y. Am I crazy?"
• "Here's my landing page copy. What's weak?"
• "Should I hire a VA or build an automation first?"

Skip the council for:

• Factual questions with one right answer
• Pure creation tasks ("write me a tweet")
• Summaries or processing tasks
• Validation-seeking when you already know the answer

The council tells you things you don't want to hear. That's the feature, not a bug.

Credit

• Methodology: Andrej Karpathy's LLM Council
• Adapted for Claude Code sub-agents by @olelehmann
• Published as an installable skill by the community

License

MIT — do whatever you want with it.