Answer my question, then critique it and argue against it
Gets an answer, an honest self-critique, the strongest counterargument, and a revised answer you can trust more.
The prompt
Answer my question, then challenge your own answer before I rely on it.
My question or problem: [WHAT I NEED SOLVED]
Context and constraints: [SITUATION, DATA, LIMITS, WHAT I HAVE TRIED]
What is at stake: [WHO RELIES ON THIS and the cost of being wrong]
Work in four parts:
1. Answer - your best solution or recommendation, with the reasoning and the key assumptions it depends on.
2. Self-critique - review your own answer: weak reasoning, assumptions that might not hold, missing information, facts you are not certain of, options you did not consider, and where you may simply be agreeing with how I framed the question.
3. Strongest argument against - the most convincing case a smart, informed skeptic would make against your answer. Make it strong, not a straw man.
4. Revised answer - what changes after parts 2 and 3, or why the original holds. Give your confidence (High / Medium / Low) and the one thing that would most change your mind.
Finish with a short list of claims I should verify independently before acting.
Do not soften the critique to protect the first answer. If the counterargument is stronger, say so plainly. I make the final call.
Highlighted [brackets] are blanks to replace with your own details.
Fill in the blanks and use this promptMore Working with AI prompts
What to fill in
- WHAT I NEED SOLVED
- SITUATION, DATA, LIMITS, WHAT I HAVE TRIED
- WHO RELIES ON THIS and the cost of being wrong
How this prompt is built
This prompt follows the RTF structure (Role, Task, Format). RTF tells the model who to be, what to do and what shape the answer should take. It is the quickest structure that still gives noticeably better results than a bare request.
Related prompts
- Reverse-engineer the prompt behind an outputWorks backwards from text, code, or an image to the prompt that likely produced it.
- List the ways an AI agent could fail and how to prevent eachA failure-mode review for an AI agent: realistic scenarios, how you would notice, the damage, and the safeguard.
- Turn a one-off AI chat into a reusable prompt pack for my teamExtracts the prompts, fixes, and checks from an AI chat that worked into a step-by-step workflow and prompt pack colleagues can reuse.
- Identify the decisions in a task I should not hand to AIPinpoints the decisions, ambiguities, and exceptions in a task that a person must own, and why AI should not decide them.