The Take
The vanity prompt is undefeated.
“What are my blind spots.” “What do I undervalue about myself.” “Describe me based on what you know about me.”
People save these. They screenshot the results.
Why do they hit so hard?
Because nobody else is paying that kind of attention.
Friends don’t ask you what you’d be great at if you took yourself seriously. Your partner might have an opinion, but it’s not an audit. Your manager isn’t analyzing your money beliefs. Even your therapist is on the clock.
AI will sit there & build a 4-paragraph read of you in 30 seconds, no eye contact required.
That’s the appeal. It’s also the risk.
The mirror is useful. Just don’t mistake attention for accuracy.
Seen This Week
1. The prompt on what AI thinks you could sell
There’s a reason everyone runs the “describe me based on what you know about me” prompts. People love hearing what AI sees in them. That’s why vanity prompts get a lot of attention.
This one uses that hook but points it somewhere useful. Instead of “describe my personality,” it asks the model to name 3 areas where you know enough that someone would pay for it.
But let’s be real. You’re not really building a product. You’re seeing which 3 things AI thinks you’re an expert in.
Based on what you know about me, identify the 3 areas where I have the most knowledge, experience, or enthusiasm. For each one, suggest a specific digital product or service I could create within 30 days. Include the target audience, a working title, and a price point. Ask for more detail if required.Why it works: “What are you good at” gets you flattery. “What could you sell” gets you a real answer. Adding a buyer, a price, and a deadline forces the model to commit, & the generic stuff dies on contact.
One gap: The output is only as good as what the model has on you. If you’ve been using it for 2 years to write business emails, it knows your writing style & nothing else. If you bounce between Claude, ChatGPT, Gemini, or whatever model you use, each one has 1/3 of the picture.
Simple fix: If you mostly use AI for surface-level tasks or you’re spread across tools, give it more to work with before running this. Paste your LinkedIn bio, a recent post, or something you wrote outside of work.
Want to go deeper: Look at the 3 ideas. Pick the one you’d write off as too basic to actually sell.
You picked [IDEA] as one of my 3. I'd write that one off as too basic. Tell me why I'm underestimating it. Be specific about who'd pay for it and why I'm wrong about the value.You’re forcing the model into a corner it wouldn’t have picked on its own, & you’re learning where your blind spot is at the same time.
2. The time audit you’d rather not see
Everyone has a story about how they spend their week.
The story is almost always wrong.
You think email takes 2 hours but it’s really 5. You think meetings eat an hour a day but they eat half of it.
The story is what you’d say in an interview. The reality is what’s on your calendar.
This prompt drops the story.
Based on what you know about me, audit how I spend my time each week. Identify the 5 biggest time drains that produce zero income or progress. For each one, calculate how many hours per week it costs me and suggest a specific money-generating activity I could replace it with. Create a revised weekly schedule that protects at least 5 hours for income-producing work.Why it works: Capping it at 5 forces the model to rank. Without the cap, you’d get a list of 12 things & end up cutting the easy ones instead of the painful ones. Requiring a replacement for each time drain turns “stop doing this” into “do this instead,” which is the difference between a guilt trip & a plan.
One gap: The model can only audit what it’s seen, so if you’ve never told it how your week actually goes, it’s guessing. The worst drains never make it onto a calendar anyway.
Simple fix: Open your calendar from the last 2 weeks. Copy the actual events into the prompt, right after “audit how I spend my time each week.” Include recurring meetings, focus blocks you didn’t actually use, & gaps you filled with email or Slack.
Then pull up Screen Time on iOS or Digital Wellbeing on Android. Add the top 3 apps by daytime use & which one you open first after each pickup.
Calendar shows where your week went on paper. Screen Time shows where it actually went.
3. The money beliefs you didn’t know you had
Most people don’t know what they actually believe about money.
They know what they say out loud (work hard, save, don’t be greedy) but those are the lines they inherited. The real beliefs run underneath.
They show up in what you charge, what you turn down, who you compare yourself to, and what you call lucky when someone else does it.
This money prompt drags those up.
Based on what you know about me, analyze my language and attitudes around money, success, and earning. Identify 3 potential limiting beliefs I might hold based on patterns in our conversations. For each belief, explain where it likely came from, how it's capping my income, and give me a specific reframe I can use daily. Then create 5 challenge exercises I can do this week to start breaking each belief down.Why it works: Asking for only 3 beliefs forces the model to commit to the heaviest ones instead of listing 10 mild ones. And asking for a daily way to flip each belief plus 5 weekly exercises closes the loop the usual “here are your patterns” prompts leave open. You get a diagnosis & a way to push back, in the same answer.
One gap: The model is delivering 3 verdicts about how you think about money with no receipts. Sounds insightful, but you have no way to argue with it because you can’t see what it’s reading from.
Simple fix: Add this to the end of the prompt:
For each belief, show me the exact lines or patterns from our past conversations you're reading this from. Then do a web search for the specific money frameworks, research, or case studies you're using to interpret those patterns as that specific belief, and cite what you find.The 1st part gives you the evidence so you can push back. The 2nd part makes the model actually look it up instead of inventing something that sounds right.
Want to go deeper: If a belief lands, you probably care less about the history of where it came from & more about how to get past it. Ask:
Take this belief: [PASTE THE BELIEF HERE]. Give me a 30-day plan for it. Include weekly milestones, what to do when I catch myself defaulting back to it, and 1 signal that tells me the belief is losing its grip.The Teardown
When the model just agrees with everything you say
You’ve felt this. You bring an idea to the model. You ask it to poke holes. It says “great question, here are 3 ways to make it even stronger.”
You didn’t get a critique. You got a polish job with the assumptions you walked in with still intact.
I came across a prompt designed to fix exactly this. The author calls it the echo chamber trap.
The instinct is sharp. AI isn’t lying to you. It’s agreeing with you in a way that sounds like analysis. The prompt tries to force the model to challenge the premise instead of polishing the execution.
Act as a critical growth strategist and cognitive auditor.
Before giving advice, analyze my idea for:
1. Unstated assumptions
What am I treating as true without evidence?
2. Confirmation bias
Where am I framing this to get agreement?
3. Hidden friction
What practical bottleneck or objection am I ignoring?
Return:
- What I said
- What might be wrong underneath
- Why it matters
- What I should verify first
End with two uncomfortable but useful questions.
Do not give me strategy yet.Prompt Teardown
The 3 categories are doing real work.
Unstated assumptions. Confirmation bias. Hidden friction.
But the prompt has 2 weak spots.
The first is the bigger one.
The prompt trusts your input completely. You paste an idea, the model runs with it, end of story.
There’s no way for the model to say “wait, before I analyze this, what’s your situation, who’s it for, what have you already tried.”
We all do this when we explain something. We have the full picture in our head & we forget the listener doesn’t. OP’s prompt has no escape for that. The model fills in whatever’s missing by guessing, then analyzes the guess.
The 2nd is the line “end with 2 uncomfortable but useful questions.” The model has no way to know what’s uncomfortable for you. It doesn’t know your blind spots, your fears, the things you’ve been avoiding. So it generates 2 generic questions that sound deep & feel like every other coaching prompt.
“Have you considered what success would actually look like.”
That’s not uncomfortable. That’s a quote from a TED talk.
The role, “critical growth strategist and cognitive auditor,” is theater. Cut it and the output doesn’t change.
Build Up
Keep the OP’s structure. Fix the 2 weak spots.
Before giving any advice, analyze the idea below across 3 failure modes:
(1) Unstated assumptions. What am I treating as fact that isn't verified? Quote the exact line you're reacting to.
(2) Framing. What word or phrase is loading the question toward a particular answer? Quote it.
(3) Friction. What's the most specific thing that would block this in the real world? Name the bottleneck, not the category.
If you need more context about my situation, what I've already tried, or who this is for, ask me before answering.
End with 2 questions that point at something I haven't told you yet, based on what's missing from my idea.
Do not give advice until I ask.
Here's the idea: [PASTE IDEA HERE].Same shape as the OP’s. Same 3 failure modes, but 3 changes.
The model has to quote what it’s reacting to instead of speaking in generalities. It can stop & ask for context before it analyzes. And the closing questions point at what’s missing from your idea instead of guessing at what’s uncomfortable.
All AI models will hide in vagueness if you let them.
Original post by u/Infamous-Ad7667. Link