Claude Prompt Library

Claude Prompts for Self-Reflection That Actually Goes Deep

17 copy-paste prompts

20 copy-paste Claude prompts for life audits, values clarification, journaling, and inner work. Claude's nuanced reasoning makes it sit closer to a thoughtful friend than a mirror that flatters.

In short: This page contains 17 copy-paste ready prompts, organized into 4 categories with a description and pro tip for each. The first 5 prompts are free instantly, no signup needed. Hand-curated and tested by the AI Academy team.

Louis Corneloup
By Louis Corneloup ยท Founder, Techpresso
Last updated ยทHand-curated & tested by the AI Academy team

Life Audit + Inventory

5 prompts

Annual Life Audit

1/17

โœจ What it does

Claude walks you through an annual life audit one question at a time across career, relationships, health, finances, learning, and joy, with a deeper follow-up before it changes topic, and tags vague versus concrete answers. Answer specifically, then stay on a dimension until you have said something real.

Walk me through an annual life audit. Ask me one question at a time across these dimensions: career, relationships, health, finances, learning, joy. After my answer, ask one follow-up that goes deeper. Don't move to the next dimension until I've actually said something specific. Tag where I'm vague vs concrete.

๐Ÿ’ก

Pro tip: One question at a time matters. Multi-question prompts get surface answers. Sequential drilling = you accidentally tell yourself something you didn't plan to say.

Energy Audit

2/17

โœจ What it does

Claude runs an energy audit on your last 7 days, sorting each activity as energizing, draining, or neutral, then names patterns and 3 changes to test next week. List what you actually did, then try one of those three changes.

Run an energy audit on my last 7 days. I'll list what I did each day. For each activity, help me categorize: gave me energy, drained me, neutral. Then identify patterns: what consistently energizes me, what consistently drains. Suggest 3 changes I could test next week.

๐Ÿ’ก

Pro tip: People mis-remember which activities drain them. Tracking 7 days then categorizing live = actual data. Often the "fun" thing drains and the "boring" thing energizes โ€” surprising patterns.

Stop-Start-Continue Self-Review

3/17

โœจ What it does

Claude sorts 5-10 patterns from the past 6 months into stop, start, and continue, and pushes back if you are too hard on yourself or too easy. Describe the real patterns, then act on one stop and one start.

Give me a stop-start-continue review of myself based on what I share. I'll describe 5-10 patterns from the past 6 months. Help me sort: what to STOP (costing more than it gives), START (overdue investments), CONTINUE (already working). Push back if I'm being too hard on myself OR letting myself off easy.

๐Ÿ’ก

Pro tip: Ask Claude to push back both ways. Without that, you get diplomatic affirmation. With it, you get a friend who notices when you're self-flagellating AND when you're hiding from a hard truth.

5-Year Reverse Look

4/17

โœจ What it does

Claude compares where you were 5 years ago with where you are now: deliberate change versus drift, what you overrated, what you underrated, surprises, and what the next 5 years look like if you stay passive. Write both snapshots honestly, then pick one deliberate change for the next stretch.

I'll describe where I was 5 years ago and where I am now. Help me see: what I changed deliberately vs what changed by drift, what I expected to matter that didn't, what I underestimated, what surprised me. Then: what does this pattern predict about the next 5 years if I stay passive?

๐Ÿ’ก

Pro tip: "Drift vs deliberate" is the question most people don't ask. Most life changes happen by accumulation, not decision. Naming which is which = power to change the ratio.

What I'm Avoiding Inventory

5/17

โœจ What it does

Claude helps you inventory what you are avoiding: postponed conversations, deferred decisions, half-known truths, and delayed work, then asks the cost of each and ranks them. Answer the questions, then face the highest-cost item first.

Help me build an inventory of what I'm currently avoiding. Ask me about: conversations I'm postponing, decisions I'm deferring, truths I half-know but don't want to confirm, work I keep finding reasons to delay. For each, ask "what's the cost of avoiding?" Then rank by avoidance cost.

๐Ÿ’ก

Pro tip: Avoidance has a cost everyone pays in anxiety + accumulated regret. Listing the avoidances often shrinks them; most are smaller than the dread around them.

Values + Identity

4 prompts

Values Clarification (Behavior-Based)

6/17

โœจ What it does

Claude infers your actual values from how you spent time, money, and energy over the last 30 days, then compares that to what you claim to value and names the gap. Describe the month as it happened, then close the loudest gap with one calendar change.

Help me identify my actual values from how I spend time + money + energy โ€” not from what I claim to value. I'll describe last 30 days. Reflect back: based on actions alone, here's what you seem to value most. Compare to what I CLAIM to value. Where's the gap?

๐Ÿ’ก

Pro tip: Stated values are aspirational. Time + money allocation = revealed values. The gap is usually where the dissatisfaction lives. Closing it (in either direction) = relief.

Identity Capture vs Identity Earned

7/17

โœจ What it does

Claude walks through identities like parent, job title, or hobbyist and asks which you chose, which captured you, which still serve you, and which you cling to after they expired. Be honest rather than gentle with yourself, then retire one identity that is done.

Walk me through my identities (parent, professional title, hobbyist, etc.). For each, ask: did I earn this through choice, or did it capture me by default? Which identities still serve me? Which have outgrown me but I still cling to? Be honest, not gentle.

๐Ÿ’ก

Pro tip: Default identities are exhausting because they're not chosen. Naming "this captured me, I didn't choose it" = first step to reclaiming agency. Some captures are still worth keeping; some aren't.

Whose Approval Audit

8/17

โœจ What it does

Claude audits whose approval you are chasing from recent decisions, posts, and things you avoid saying, and asks whose voice is in your head and whether they are even watching. Name the voice, then make the next choice as if that audience were gone.

Help me audit whose approval I'm subconsciously chasing. I'll describe recent decisions, what I post online, what I avoid saying. Reflect: whose voice is in my head when I make these choices? Are those people I respect? Are they even paying attention? Often we perform for an audience that isn't watching.

๐Ÿ’ก

Pro tip: 90% of the audience you're performing for isn't paying attention. Naming the specific people = realizing how few of them you actually respect = freedom.

Eulogy Exercise

9/17

โœจ What it does

Claude walks you through a eulogy for 50 more years lived on your values: who speaks, three things they say you were, the actions that proved it, what you taught, and what would be missing on your current path. Refuse a generic draft, then change one action so the missing piece appears.

Walk me through writing my eulogy as if I lived 50 more years living true to my values. Ask: who's speaking, what 3 things they'd say I was, what specific actions justified that, what they'd say I taught them, what would be conspicuously absent (compared to current trajectory). Don't let me write a generic eulogy.

๐Ÿ’ก

Pro tip: Generic eulogies = "kind, hardworking, loved family." Useless. Specific eulogies = "shipped weird projects, took 3 sabbaticals, said no often." That's direction.

Decisions + Direction

4 prompts

Big Decision Pre-Mortem

10/17

โœจ What it does

Claude runs a pre-mortem on [decision]: assume it failed badly in 2 years, then assume it succeeded beyond hope, and compares the two so you can see the imbalance. Answer both futures, then let that imbalance steer the choice.

I'm considering [decision]. Walk me through a pre-mortem: assume I made this choice and 2 years later it failed badly. What went wrong? Then assume it succeeded beyond expectations โ€” what made the difference? Help me see the asymmetry between scenarios. What does that suggest?

๐Ÿ’ก

Pro tip: Imagining failure paths surfaces risks the optimistic mind hides. Imagining wild success surfaces conditions you'd need to engineer. Pre-mortem > pros/cons list.

The "If Nobody Knew" Filter

11/17

โœจ What it does

Claude runs the if-nobody-knew filter on [option A] versus [option B]: which you would pick if no one ever found out, and whether that differs from the public answer. Notice the gap, then treat it as status talking if the private pick is different.

I'm torn between [option A] and [option B]. Run the "if nobody knew" filter: which would I choose if no one would ever find out which I picked? Help me notice if my answer differs from what I'd announce publicly. The gap between private + public answer = where status motivation is steering.

๐Ÿ’ก

Pro tip: Status drives more decisions than people admit. The "if nobody knew" filter strips it. Often the answer flips โ€” revealing the public choice was for an audience, not for you.

Regret Forecast

12/17

โœจ What it does

Claude asks, at the fork [describe options], which choice you would regret making and which you would regret not making, so visible failure and the road not taken both get weight. Answer both regrets, then choose with that balance in mind.

I'm at a fork: [describe options]. Ask me: which option am I more likely to regret choosing? Which am I more likely to regret NOT choosing? People over-weight visible failure regret + under-weight invisible road-not-taken regret. Help me balance both.

๐Ÿ’ก

Pro tip: Daniel Kahneman: people regret action more than inaction in the short term, but inaction more than action in the long term. Long view = the road not taken haunts longer than the failed attempt.

Walk-Away Test

13/17

โœจ What it does

Claude tests whether walking away from [situation] is temporary friction or real misalignment, and asks what would have to change for you to stay happily, plus whether you have said those conditions out loud. If you cannot name conditions, treat that as data; if you can, say them.

I'm considering walking away from [situation]. Help me test: is this dissatisfaction temporary friction (will pass) or signal (real misalignment)? Ask me what would have to change for me to stay happily. If I can't name conditions, that's data. If I can name them, ask: have I stated them?

๐Ÿ’ก

Pro tip: "What would have to change?" is the diagnostic. If the list is short + doable + un-asked-for, it's communication failure, not incompatibility. If the list is unfixable, it's signal.

These prompts give you the what. Tutorials give you the why.

Learn when to use extended thinking, how to build Claude Projects, and workflows that compound. 300+ tutorials and growing.

Try AI Academy Free

Inner Work + Reflection

4 prompts

Pattern Recognition from Journal

14/17

โœจ What it does

Claude finds unnoticed patterns in [Paste journal entries from last 30-90 days]: recurring themes, triggers, loops you write about without resolving, and contradictions, quoting you back to yourself. Paste real entries, then sit with the quote that repeats.

[Paste journal entries from last 30-90 days]. Find patterns I haven't noticed: recurring themes, emotional triggers, things I write about repeatedly without resolving, contradictions in what I claim to want. Reflect them back specifically โ€” quote me back to myself.

๐Ÿ’ก

Pro tip: Claude's long context lets you paste 10+ journal entries. Patterns visible across entries that aren't visible day-by-day. Often someone's been writing the same problem for 6 months without seeing it.

Younger-Self Letter

15/17

โœจ What it does

Claude helps you write a letter to your [age] self: warnings, encouragement, what they could not see, apologies, and thanks for surviving, pushed toward specifics. Fill the age, then keep only the lines that could only be about you.

Help me write a letter to my [age] self. Walk me through it: what I'd warn them about, what I'd encourage, what I now know they couldn't see, what I'd apologize for, what I'd thank them for surviving. Make me specific โ€” generic letters are useless.

๐Ÿ’ก

Pro tip: The letter to younger self often surfaces what you needed to hear FROM SOMEONE ELSE at that age. That's the thing you should still tell yourself now โ€” the wound didn't close just because you got older.

Triggered Reaction Decoding

16/17

โœจ What it does

Claude decodes a triggered reaction after [event] when you felt [emotion] out of proportion: the story you told, the meaning you assigned, where that pattern started, and what it is really about. Answer the questions, then separate the surface event from the older pattern.

Yesterday [event] triggered a strong reaction in me โ€” I felt [emotion] disproportionate to the situation. Help me decode: what story am I telling about what happened? What's the meaning I'm assigning? Where did this meaning-making pattern come from? What's the pattern actually about (often not the surface event)?

๐Ÿ’ก

Pro tip: Disproportionate reactions = old wound activated, not current event sized. The work isn't suppressing the reaction; it's naming what the trigger is actually about so you stop reacting to ghosts.

Permission Slip

17/17

โœจ What it does

Claude walks you through what you are waiting for permission to do, who you are waiting on, why their yes feels required, and whether you can grant it yourself, and will write a literal permission slip if that helps. Write the slip, then do the thing without waiting.

What am I waiting for permission to do? Walk me through it: what I want to do, who I'm unconsciously waiting on permission from, why their permission feels needed, whether I can grant it to myself. Help me write a literal permission slip if useful.

๐Ÿ’ก

Pro tip: Adults still wait for permission they don't need from people who aren't holding it. Naming the specific permission + the specific person = realizing both are imaginary. Then granting yourself the slip.

Free tool

Prompt Optimizer

Turn a rough idea into a structured, professional AI prompt.

Try it free โ†’

Frequently Asked Questions

They're not substitutes. Claude = 24/7 thinking partner for self-reflection, structured exercises, pattern recognition. Therapist = trained professional for clinical issues, trauma, mental health conditions. Use Claude for inner work; use therapy for actual treatment.
Less than ChatGPT, but still some. Add "push back if I'm self-deceiving" or "be honest, not encouraging" to your prompts. Claude responds to the permission you grant. Default mode tilts diplomatic; explicit license unlocks honesty.
Less weird than people think. Pen-and-paper journaling has 1 advantage: privacy. AI journaling has 2: it asks follow-ups + finds patterns across entries. Different tools; both useful. Many people use both.
Anthropic's default consumer Claude doesn't train on your conversations (per current policy โ€” verify). For maximum privacy, use Claude API with no logging. For deeply private content, paper journaling still the gold standard.
Partially. For structured exercises (values clarification, decision frameworks, journaling prompts), yes. For accountability + pattern-spotting across long timelines + human warmth, no. Best stack: Claude for daily reflection, human coach for monthly depth work.

Prompts are the starting line. Tutorials are the finish.

A growing library of 300+ hands-on tutorials on ChatGPT, Claude, Midjourney, and 50+ AI tools. New tutorials added every week.

7-day free trial. Cancel anytime.