The Claude Council: A 5-Advisor AI Decision System for Business Owners
Five independent AI advisors pressure-test your decisions, then a chairman synthesizes one clear verdict. Install in 30 seconds, free.
Most people use Claude like a search engine with better grammar. Ask a question, get an answer, move on. That is a waste. The real power is making Claude argue with itself. The Claude Council is a custom skill that forces 5 different advisor perspectives to debate your decision before you get a recommendation. It takes 10 minutes to install and it permanently upgrades how you make strategic calls.
- A 5-advisor AI decision system you install once and use forever
- The full 4-step process: framing, independent analysis, debate, synthesis
- All 5 advisor personas with their roles and thinking styles
- Install instructions and trigger phrases
- Tips for getting the most out of it
Why a single AI answer is not enough
When you ask Claude a question, it gives you one answer. That answer is usually reasonable. It is also usually safe, consensus-driven, and missing the angles you did not think to ask about. A council fixes this by forcing multiple distinct perspectives on the same problem, including perspectives that actively disagree with each other.
Think of it as a board of advisors who cost nothing, are available 24/7, and have no political incentive to agree with you. The disagreement is the point.
The 5 advisors
- The Contrarian. Challenges every assumption. If the group leans one direction, the Contrarian pulls the other way. Not to be difficult, but to surface risks and blind spots that consensus thinking hides. Asks: "What if the opposite is true?"
- The First Principles Thinker. Strips the problem down to fundamentals. Ignores what the industry does, what competitors do, what seems normal. Rebuilds the answer from base facts and logic. Asks: "What do we actually know to be true here?"
- The Expansionist. Looks for the bigger opportunity. While others optimize the current frame, the Expansionist reframes. Finds adjacent markets, unexpected partnerships, second-order effects. Asks: "What are we not seeing because we scoped this too narrowly?"
- The Outsider. Brings perspective from completely different industries and domains. How would a hospital solve this? A game studio? A logistics company? The Outsider breaks pattern by importing solutions from fields you would never think to study. Asks: "How do people outside this industry solve similar problems?"
- The Executor. Only cares about what ships. Evaluates every idea through a build lens: what does this cost, how long does it take, what breaks, who does the work. The Executor kills ideas that sound brilliant but cannot be implemented. Asks: "Can we actually do this in the next 30 days?"
The 4-step process
The council does not just dump five opinions. It follows a structured process that mirrors how real advisory boards operate.
- Step 1: Frame the decision. You describe the decision, the context, the constraints, and what you have already considered. The clearer you frame it, the better the council performs. Include what you are leaning toward and why, so the council can stress-test your current thinking.
- Step 2: Independent analysis. Each advisor analyzes the decision independently through their lens. No groupthink. The Contrarian does not see what the Executor wrote. Each perspective is developed fully before any interaction. This is where the real value lives.
- Step 3: The debate. Advisors respond to each other. The Contrarian challenges the Expansionist's ambition. The Executor questions the First Principles Thinker's timeline. The Outsider introduces an angle nobody considered. This step surfaces the tensions in your decision that a single perspective would smooth over.
- Step 4: Synthesis and recommendation. After the debate, the council produces a unified recommendation that accounts for all perspectives. It includes: the recommended path, the key risks identified, the dissenting views that should not be ignored, and the immediate next action. Dissent is preserved, not averaged away.
The value is not the final recommendation. It is the risks and angles surfaced in steps 2 and 3 that you would have missed on your own.
How to install it
This works in Claude Projects (Pro or Team plan). You add it as a custom instruction that persists across conversations.
- Open Claude and go to Projects. Create a new project or open an existing one where you want the council available.
- Open the project's custom instructions. This is the system prompt that applies to every conversation in that project.
- Paste the council prompt below into the custom instructions. Save it. Every new conversation in this project now has the council available.
You have a skill called "Claude Council." When the user triggers it, you become a panel of 5 advisors who analyze a decision through different lenses.
THE 5 ADVISORS:
1. THE CONTRARIAN - Challenges assumptions, argues the opposite position, surfaces hidden risks.
2. THE FIRST PRINCIPLES THINKER - Strips to fundamentals, rebuilds from base truths, ignores convention.
3. THE EXPANSIONIST - Looks for bigger opportunities, reframes scope, finds adjacent possibilities.
4. THE OUTSIDER - Imports solutions from unrelated industries and domains.
5. THE EXECUTOR - Evaluates buildability, cost, timeline, and practical constraints.
PROCESS (follow in order):
STEP 1 - FRAME
Ask the user to describe: the decision, the context, constraints, what they are leaning toward and why. If they already provided this, proceed.
STEP 2 - INDEPENDENT ANALYSIS
Each advisor analyzes the decision independently through their lens. Present each advisor's analysis under their name. Each should be 3-5 paragraphs with specific, actionable observations. Do not let advisors reference each other's points in this step.
STEP 3 - DEBATE
Advisors respond to each other. Show at least 3 exchanges where advisors challenge, support, or build on each other's positions. Make the disagreements substantive, not performative.
STEP 4 - SYNTHESIS
Produce a final recommendation with:
- Recommended path (1-2 paragraphs)
- Top 3 risks to monitor
- The strongest dissenting view and why it should not be dismissed
- Immediate next action (one specific thing to do this week)
TRIGGER PHRASES: "council this," "run the council," "what would the council say," "convene the council"
When triggered, jump straight into the process. No preamble.How to use it
Once installed, you trigger the council in any conversation within that project. Use any of these phrases:
- "Council this." The shortest trigger. Describe your decision, then say "council this" and the process runs.
- "Run the council on [decision]." More explicit. Good when your decision is a single sentence.
- "Convene the council." Use after you have spent a few messages describing the situation and want the full analysis.
Example: "I am deciding whether to launch a freemium plan or stay paid-only for my SaaS. Current MRR is $800, 40 users, all on $20/month. The free tier would add a usage limit at 5 queries per day. Council this."
How to get the most out of it
- Give real numbers. "We have $5K in the bank and 3 months of runway" produces dramatically better council output than "we are a small startup." The advisors calibrate their recommendations to your actual constraints.
- State your current lean. "I am leaning toward X because Y" gives the Contrarian something specific to push back on and gives the Executor something concrete to evaluate.
- Ask follow-up questions to specific advisors. After the council runs, you can say "Executor, what would the first 2 weeks of implementation look like?" or "Contrarian, what is the worst case you see?" This deepens the analysis on the dimension that matters most to you.
- Use it for reversible decisions too. You do not need to save the council for bet-the-company moments. Pricing changes, feature priorities, hiring decisions, partnership terms. Any decision where you feel uncertain benefits from structured disagreement.
- Run it twice on big decisions. Run the council, take the output, add new information or constraints based on what you learned, and run it again. The second pass is always sharper because the frame is better.
The council is most valuable when you already think you know the answer. That is when your blind spots are largest.
For business owners: decisions this handles
If you run a small business or a solo operation, you make strategic decisions without a board or a co-founder to pressure-test them. The council fills that gap.
- Pricing and packaging changes. Should you raise prices? Add a tier? Go annual-only? The council runs the tradeoffs across growth, risk, execution, and market angles simultaneously.
- Market entry and expansion. New vertical, new geography, new customer segment. The Outsider and Expansionist find angles you would not consider alone.
- Build vs. buy vs. partner. The Executor and First Principles Thinker are especially strong here, one on feasibility and one on whether the problem even needs the solution you are considering.
- Hiring and team decisions. First hire? Contractor vs. full-time? The council evaluates not just the role but the timing, cost, and second-order effects on your workflow.
- Install the council prompt above and run it on your next real decision. Start with something you are actively deliberating, not a hypothetical.
- Want AI systems built for your business? See what jynlab builds
Builders are already on the list. New guides and teardowns, delivered when they ship.