actionable ai
FREE HOW-TO GUIDE · FROM THE AI OPERATOR VAULT
Set Up Your Own Claude Council

Free guide · by Cooper Simson

Set Up Your Own Claude Council

Full step-by-step guide to set up your own Claude Council, plus my top use cases and prompts for battle-testing every decision.

Get the guide, free

No email required. Use it today.

Claude has a major problem that no one talks about: it is the world's best yes man. It does not matter if your idea is good or terrible, Claude finds a way to agree with you. The Claude Council skill fixes that by battle testing every real decision with 5 independent advisors.

What You'll Learn

  • The full step by step install for the Claude Council skill
  • How 5 independent advisors battle test your questions
  • Top use cases: product decisions, features, brand deals, pressure testing
  • Copy paste prompts to council your first real decision

The Problem

You ask Claude "is this a good idea?" and it builds a case for why it is. Give it the same idea and ask "why is this bad?" and it builds an equally convincing case in the other direction. One perspective is not enough for real decisions. This is why you can never fully trust a single Claude response for choices that actually matter.

Step By Step Install

  1. Get the skill file. Copy the full Claude Council skill file from Ole Lehmann's Google Doc. It is modeled on Andrej Karpathy's LLM Council methodology.
  2. Create the skill directory. Open your terminal and run:
mkdir -p ~/.claude/skills/llm-council
  1. Save the skill file. Paste the skill content into a file called SKILL.md inside the directory you just created:
~/.claude/skills/llm-council/SKILL.md
  1. Test it. Open Claude Code in any project and type:
council this: [your question]

If the 5 advisors start spinning up, you are good to go.

How The Council Works

When you type "council this" followed by your question, Claude battle tests it with 5 independent advisors who each think from a completely different angle:

  • One hunts for fatal flaws
  • One strips away your assumptions and asks what you are actually trying to solve
  • One looks for the upside you are missing
  • One gets zero context about you and catches your blind spots
  • One only cares about what you should do next

Each advisor gives their independent take. Then all 5 responses are anonymized and reviewed by 5 peer reviewers who do not know which advisor said what. Finally, a chairman agent synthesizes everything into one final recommendation, telling you where the advisors agree, where they clash, and what you should actually do. You get a clear verdict with one concrete next step, plus a full HTML report saved to your workspace.

Top Use Cases

  1. Product and code decisions. Before committing to an architecture, a pricing model, or a product direction, run it through the council. The Contrarian advisor alone can save you from bad calls.
  2. Which features to build vs skip. When you have bandwidth for one feature and three options, the council pressure tests each one. The Executor advisor usually settles it by asking "what can you actually ship this week?"
  3. Whether a brand deal is worth taking. Paste the brand deal details and ask the council if it is worth your time. The Outsider advisor is great for this because it does not know your history and judges purely on what is in front of it.
  4. Pressure testing ideas before you commit. Any time you are about to invest real time into something, council it first. If the council comes back split, dig deeper. If 4 out of 5 agree, move fast.

Copy Paste Prompts

Pricing decision

council this: I'm thinking of pricing my [product] at $[X]. My audience is
[audience]. Is this the right price point or am I leaving money on the table?

Feature priority

council this: I have time to build one feature this week. Option A is [feature].
Option B is [feature]. Which one moves the needle more for [goal]?

Brand deal evaluation

council this: I got offered a brand deal from [company]. They want [deliverable]
for $[amount]. Here's the context: [details]. Is this worth my time?

Product pivot

council this: I'm thinking of pivoting from [current thing] to [new thing]. Here's
my reasoning: [context]. Am I crazy or is this the right move?

Landing page / messaging review

council this: here's my landing page copy for [product]. What's weak? What would
make someone leave without buying? [paste copy]

When To Use It

The council is for decisions where being wrong is expensive. If you already know the answer and just want validation, the council will tell you things you do not want to hear. That is the point.

  • Do not council simple questions with one right answer
  • Do not council creation tasks like writing a tweet
  • Save it for real decisions with real stakes

Once you install it you cannot live without it.

Go Deeper

Prefer a printable version? Grab the PDF: https://drive.google.com/file/d/17QEZdgevSeXMiEuBk7qIOIRhzVqkwvmA/view?usp=drivesdk

Want the whole system installed live, with Cooper on the call? Build Your Own AI, the two week sprint runs the first two weeks of August.

Or get a new working tool every Monday inside The AI Operator community on Whop.

The honest part

The guides are free. The working agents are $97.

Cooper has published 125+ guides like this one, free, no email gate. The step after reading is running the actual systems: five working agents, including the content engine behind his 10M+ views, plus your first month inside The AI Operator community, a new working tool every Monday.

Want it built with you, live? Build Your Own AI · $1,997 · live Aug 3