Cindy Zhu.
Free guides
Productivity & thinking Claude

Ask Claude what it knows about you

Hey, it's Cindy ๐ŸŒฑ Here is the full prompt list from the reel: the four in order, four more for when you want to go deeper.
btw the waitlist for AI System Club is now open! If you're keen on using AI to automate your content, video editing, job search, business or build AI agents, feel free to sign up (only if you want to hehe). Ps. the waitlist gets first dibs:)
before you start

Set it up so the answers are real

The reason most people get a vague horoscope back is that Claude has nothing to work from. Two features decide how good this gets, and they are not the same thing, so it is worth 90 seconds to check both.

๐Ÿง  Memory Claude builds a summary of you from your chats: your role, projects, working style, preferences. Available on Free, Pro and Max. It is a synthesis, not a transcript, so it holds the gist rather than your exact words.
๐Ÿ”Ž Search past chats Claude goes and reads your actual earlier conversations when you ask. This one is on paid plans only (Pro, Max, Team and Enterprise), and you will see it run as a tool call in the chat.
  1. ๐Ÿง  Turn memory on. Open Claude, go to Settings, then Memory, and switch on "Generate memory from chats". If your account is still on the older layout, go to Settings, then Capabilities, and find Memory in the Preferences section.
  2. ๐Ÿ‘€ Read what is already there. The same Memory panel shows everything Claude currently remembers, grouped by category. Read it before you run the prompts. Half the shock of this exercise is right there in plain text, and you can delete or correct anything that is wrong.
  3. ๐Ÿ”Ž Check search past chats. On a paid plan it is on by default, with a "Search and reference chats" toggle in the same Memory settings. On Free it is not available, so use the fallback below.
  4. ๐Ÿ’Œ On Free, feed it the context yourself. Start one fresh chat and paste in five to ten paragraphs about your last year: what you worked on, what you kept putting off, what you were anxious about, what you were proud of. The prompts work just as well, they just read your paste instead of your history.
  5. ๐Ÿ“ Know the gaps. Projects keep their own separate memory, so work you did inside a project will not show up in a general chat. Anything you ran in an incognito chat was never stored at all.
Nothing here reaches outside your own account. Claude is reading conversations you had with Claude, not your email, your camera roll or the internet's opinion of you. The reason it feels uncanny is that you told it all of this yourself, a bit at a time, across months.

prompt 1

Tell me everything you know about me

Start a new chat rather than continuing an old one, so the answer is drawn from your history overall instead of whatever you were last talking about. Run all four prompts in the same chat, in order, because each one builds on the last.

๐Ÿง  the full picture
Using your memory of me and everything in our past conversations, tell me everything you know about me. If you can search my earlier chats, do that first and tell me roughly how far back you looked.

Cover my personality and how I actually behave rather than how I describe myself, what I seem to genuinely want out of my life and work, what I am most afraid of, what I keep circling back to, and where my stated goals and my actual behaviour do not match. Include the small, specific details, because those are the ones that will tell me whether you are really reading my history or generalising.

Write it as if you were briefing someone who is about to work closely with me and needs to know what I am like to deal with, not as a flattering summary. Be specific and be direct, and where you are inferring rather than recalling, mark that inference clearly so I know which parts are evidence and which parts are guesses.
prompt 2

What are my blind spots?

๐Ÿ”Ž patterns I cannot see
Staying in this same chat, go deeper. What patterns have you noticed in how I think, decide and avoid things that I probably cannot see in myself?

Look specifically for repetition: problems I have described more than once in different words, decisions I keep deferring, the kind of task I always find a reason to postpone, the way I talk about myself when something has not worked, and any gap between the standard I hold other people to and the one I hold myself to. Quote or paraphrase the moments in our history that led you to each conclusion, so I can check your reasoning instead of just believing you.

Give me the three most useful ones rather than a long list, and put the one I am most likely to argue with first. Do not soften them to protect my feelings, and do not add reassurance at the end. If the evidence is thin for one of them, say so rather than padding it out.
The instruction to quote its evidence is the part that makes this worth doing. Without it you get plausible-sounding personality writing. With it, you can tell the difference between a real pattern in your history and a nice sentence.
prompt 3

What is holding me back, and where does it lead?

๐ŸŒฑ the five-year version if nothing changes
Based on everything above, what is the single biggest thing holding me back right now? Name one thing, not a list, and explain why you picked that one over the other candidates.

Then describe, specifically, what my life and work look like in five years if I never address it. Do not write a scary abstract warning. Walk me through it concretely: what my typical week looks like, what I am doing for work, what I have and have not built by then, what I am likely telling myself about why, and what the second-order costs are on the parts of my life I care about outside work.

Ground it in what you actually know about my situation, my field and my current trajectory rather than generic life advice. Where you are extrapolating, say what assumption you are making, so I can tell you if it is wrong.
prompt 4

Now flip it

๐Ÿ’ก the five-year version if I fix it
Now flip it. If I fix that one thing, starting this month, what do those same five years look like instead? Be specific and concrete in the same way you were about the bad version, with the same structure, so I can hold the two side by side.

Then work backwards from that picture. Give me the first 90 days as three monthly milestones, each with one measurable outcome, and tell me what has to be true at the end of month one for the rest to be realistic. Keep it to things I can start without permission, budget or anyone else's agreement.

Finally, name the most likely way I will quietly abandon this, based on the patterns you identified earlier in this chat, and tell me the specific early warning sign to watch for and what to do the week it shows up.

go deeper

4 more prompts for the same chat

Run these in the same conversation once the first four are done. They reuse everything Claude has already worked out about you, so they get sharper answers than they would cold.

๐Ÿง  my actual strengths, with evidence
Ignore what I have claimed about myself and tell me what I am genuinely good at, based only on things you have seen me actually do in our conversations. For each strength, give me the specific evidence you are drawing on rather than a compliment.

Then tell me which of those strengths I am currently underusing, and where in my work or life it would create the most value if I leaned into it deliberately. Rank them by that, not by how strong they are.

Finish by naming one strength I over-rely on to the point that it has become a weakness, and describe what that looks like in practice when it happens.
๐Ÿ”Ž the story I keep telling
Across our conversations, what story do I tell about myself most often? I mean the recurring explanation I give for why things are the way they are: about my work, my circumstances, my personality, or my timing.

Tell me where that story is accurate and genuinely useful, and where it has become a way of avoiding a decision. Use examples from our history rather than speaking generally, and quote my own framing back to me where it is revealing.

Then rewrite that story in one paragraph, keeping every fact the same, in a version that would leave me with more room to act. Show me both side by side.
๐Ÿ’ก what I should stop doing
Based on what you know about my week, my goals and my energy, what should I stop doing? Give me five things, ranked by how much they cost me relative to what they return.

For each one, tell me what I am really getting out of it, because I would not keep doing it otherwise: whether it is comfort, avoidance, identity, other people's expectations, or genuine but small value. Be specific about which.

Then pick the single one I am most attached to and make the strongest possible case for dropping it, including what I would need to put in its place so the gap does not just refill with something similar.
๐ŸŒฑ the letter from five years ahead
Write me a short letter from the version of me who is five years into the good scenario we mapped out. Write it in my voice, using how I actually talk in our conversations, not in a generic inspirational tone.

The letter should tell me what that version of me is glad I started doing this month, what she wishes I had stopped worrying about sooner, and the one decision from this year that turned out to matter far more than it seemed at the time.

Keep it under 300 words, keep it specific to my real circumstances, and do not make it saccharine. It should read like a note from someone who knows exactly how hard the boring middle part was.

worth knowing

The honest bit

  • Memory is a summary, not a recording. Claude stores a synthesis of your context rather than your full transcripts, so it can be confidently wrong about details. Read the Memory panel and correct anything off before you take the answers seriously.
  • It is fluent, which is not the same as insightful. A well-written paragraph about your fears can be pattern-matched from how you write rather than from what you have actually said. The "quote your evidence" instruction in these prompts exists precisely to separate the two.
  • Ask it to disagree with itself. Once you have the answers, a single line does a lot of work: "argue the opposite of what you just told me, as convincingly as you can." Anything that survives that is worth acting on.
  • You control what it keeps. You can view, edit and delete memory entries, export or import your memory, or run a conversation in incognito so nothing is stored at all.
  • It is a mirror, not a professional. This is a reflection exercise and a good one, and it is not therapy or a diagnosis. If something heavy surfaces, that is worth taking to a real person who knows you, not a longer prompt.
Save the chat and put a reminder in your calendar for three months. Run prompt one again then and compare the two answers. Watching what changed in how you describe your own year is often more useful than the original reading was.

Proof

Chat search and memory, how they work and which plans get them (Claude Help Center)
Import and export your memory from Claude
Use incognito chats
Anthropic on bringing memory to Claude

Want to use Claude like this every day, not once? Inside AI System Club I teach people to learn AI from scratch and build real systems for their work and their own business. Doors open soon and founding pricing dies at close. ๐ŸŒฑ Join the waitlist (link in my bio too).
the waitlist is now open

AI System Club is launching soon

Build AI systems with me through recorded lessons you can follow at your own pace. Join the waitlist for first access and the founding price.

btw this list gets first dibs
the moment doors open :)