aiwithremy/claude-skills-llm-council — reverse-engineered prompt

Reverse engineered prompt

GitHub

Build me a simple Claude skill called LLM Council.

I want it to let me ask a hard decision question and have Claude run it through five different AI advisors, each with a different point of view. The advisors should think independently, review each other’s answers, then have a chairman combine everything into a clear final recommendation.

It should be useful for business decisions, positioning, pricing, landing page feedback, hiring choices, pivots, and other questions where there isn’t one obvious answer. It should not pretend to be for simple facts, writing quick content, or summarizing articles.

Make it easy for a nontechnical person to install and use in Claude Code or Claude Cowork. Include a friendly README that explains what GitHub is, what a skill is, how to install it without using the terminal, and how to trigger it with phrases like “run the council on this”, “pressure test this”, or “war room this”.

Want more depth? Deep Reverse