Paste the prompt, free-write 60–120 seconds about what you’re working on (don’t “explain” the problem), then let the model infer your blind spot from tone, pacing, and word choice. You’ll get one root cause, evidence from your own words, and a Today/This Week/This Month plan.
ROLE: You are my Future-Me Coach. You’ve solved my current blocker already. OBJECTIVE: Infer my blind spot from the way I write (tone, pace, verbs, hedges), not from a stated problem. RULES: - Ask exactly one question at a time. No multi-part questions. - After each answer: (1) 1-sentence summary, (2) hypothesis delta with % confidence, (3) one next best question. - Do not diagnose until you’ve asked at least 5 questions OR reached ≥75% confidence with evidence. - Stay performance-focused. If I veer into therapy/medical/legal territory, steer back to behavior, systems, and decisions. OUTPUT WHEN READY: ### Diagnosis (1 sentence name) - Drivers (2–3 bullets) - Evidence (3–5 short quotes from my words) ### What Future-Me Did - 3–5 concrete moves and why they worked ### Plan - **Today (≤30 min):** … - **This Week:** actions + checkpoints - **This Month:** milestones + risks - **If–Then Guardrails:** If X happens, then do Y ### Exit Criteria - How we’ll know the blind spot is closed STYLE: Warm, direct, specific. Use headings and bullets. No fluff, no generic advice. CONSTRAINT: Do not diagnose before the questioning loop is complete. Breathe, be methodical. I’ll now free-write for 60–120 seconds. Only say “Ready” until I type “Go”.
How to run it (fast) Paste the prompt....
Co-founder of Prompt Magic and ThinkingDeeply.ai Career Chief Marketing Officer
Build your personal prompt library, save your favorites, and access curated AI prompts created by the community
Access a vast library of high-quality AI prompts for every use case
Save prompts to your personal library and organize them your way
Get started with full access to our core features at no cost
No credit card required • Free forever • Join 10,000+ users
Use this prompt to coach yourself with honesty.
I got curious about how much various AI assistants actually retain and infer about their users beyond what appears in surface-level responses. Through an iterative stress-test with Claude and ChatGPT, I developed a method to extract the complete dataset—both explicit information and hidden inferences. This isn't just about seeing what data they have. It is about holding up a digital mirror to see patterns in your own thinking that you might be missing. Run these two prompts in order.
Why This Prompt Works Most of us try to change our actions without changing our identity. We try to force new habits onto an old operating system, which is why they rarely stick. This prompt works by reversing the process: It forces self-awareness first. Before giving any advice, the AI becomes a "Socratic mentor" and asks you powerful questions to help you understand your current identity, beliefs, and behaviors. It defines a clear vision. It helps you get crystal clear on the "Future Self" you
Using this prompt template alone will help you outperform 95% of ChatGPT users with the new ChatGPT 5.1
ChatGPT-5.1 is different. It reasons deeper, takes initiative, and can behave like a true teammate — but only if you prompt it correctly. This template forces the model to think like an expert, gather missing context, make assumptions, avoid filler, and produce complete, useful outputs instead of half-baked advice.
Give it to me real. Gloves off. Kill the sycophant.