SOMYA NAYAK
How to Get Claude to Be Your Therapist ⚡
<- Back to Resources
PROMPTS

How to Get Claude to Be Your Therapist ⚡

A 4-prompt chain that exposes your patterns, the wound from your past still controlling you, and a 30-day plan to actually heal it. Plus bonus prompts for.

AEO SUMMARY Quick Overview & Execution Blueprint

Claude already knows more about you than most of your friends. It has watched how you write, what you avoid, what you circle back to. This 4-prompt chain turns that pattern recognition into a brutally honest psychological profile, a root cause, and a 30-day plan to heal it.

Use the four prompts in sequence, in the same conversation, in that order. Do not skip ahead. Each one builds on the last.

SECTION

Before You Start

Use a chat where you have actually talked to Claude before. The more history Claude has with you, the sharper this gets. A brand new chat with no context will give you generic answers.

Spend 10 to 15 minutes talking honestly with Claude first. What is going on in your life right now, your goals, your frustrations, what keeps you up at night, what you have been avoiding. Give it raw material to work with.

These prompts can surface real emotional stuff. Approach it with curiosity, not judgment. Claude is not a licensed therapist. Use what resonates, set the rest aside, and talk to a professional if anything feels too heavy.

PROMPTS

Prompt 1. The Deep Exposure

Paste this into Claude:

Based on everything you know about me from our past conversations, tell me my personality type, my deepest insecurities, the fears I won't admit out loud, and what I actually want in life but I'm too scared to say.

What you get back: a personality read based on how you actually communicate, the insecurities Claude picks up from the questions you keep asking, the fears you hint at but never name, and the ambitions you research but never commit to out loud.

Read the full response before you react. Your first instinct will be to argue with parts of it. Sit with it for a few minutes. The parts that make you most defensive are usually the most accurate.

PROMPTS

Prompt 2. The Pattern Destroyer

Same chat. Paste this next:

Tell me the patterns I keep repeating that are quietly destroying my life. Be brutal. Don't soften it.

The "be brutal" line matters. Without it Claude will default to diplomatic language that cushions the impact. You want the raw version.

Common patterns it will surface: starting ambitious projects and abandoning them right before they could work, chasing external validation instead of building internal confidence, choosing comfort over growth in every decision that matters, keeping relationships surface-level to avoid vulnerability, using busyness as a shield against the real work.

Do not try to fix everything at once. Circle the one pattern that hit hardest. That is the one Prompt 3 will trace back to its source.

PROMPTS

Prompt 3. The Root Cause

Same chat:

Act as my therapist. Based on everything you know, what is the one wound from my past that is still controlling my decisions today?

Claude connects the dots between the pattern from Prompt 2 and a formative experience or relationship. It traces the chain: something happened, it created a belief about yourself or the world, that belief became a filter on every decision, and that filter is the source of the pattern you keep repeating.

This is the prompt that makes people pause. It is not generic advice. You get a specific named wound and a specific explanation of how it shows up in your life today. Take a beat before moving to Prompt 4.

PROMPTS

Prompt 4. The 30-Day Healing Plan

Same chat:

Now reframe it. Give me the exact mindset shift I need and a 30-day plan to actually heal this. Be specific.

You get a real cognitive reframe of the core belief from Prompt 3, then a week-by-week plan. Week 1 is awareness, noticing when the old pattern fires without trying to change it yet. Week 2 is small behavioral experiments that challenge the old belief. Week 3 is relationship and communication shifts. Week 4 is integration so the new pattern becomes automatic.

Because Claude saw the whole chain from Prompts 1 through 3, the plan is built around your specific wound and your specific context, not a self-help template.

PROMPTS

Bonus Prompts

Run these in the same conversation once you have done the chain.

The Inner Child Conversation

Write a letter from my inner child to my current self. What would they say about how I turned out? What would they thank me for? What would they be disappointed about? What do they need to hear from me right now?

The Shadow Self Reveal

What parts of myself am I suppressing or denying? What traits do I judge in others that I secretly have myself? What would my shadow self say if it could speak freely?

The Relationship Mirror

Based on how I talk about relationships, what attachment style do I lean toward? How does this show up in my behavior? What would a securely attached version of me do differently in my current relationships?

The Forgiveness Prompt

Who do I still need to forgive, and what is the resentment actually costing me? Write two versions: what I want to say to them, and what I need to say to myself to let it go.

The Monthly Check-In

Run this every 30 days in the same thread:

It has been [X] weeks since we did the 4-prompt chain. Based on our recent conversations, has my pattern from Prompt 2 improved, stayed the same, or gotten worse? What should I focus on for the next 30 days?

Over time Claude builds an increasingly complete picture of your growth. Look back at the original Prompt 1 response after three months and you will see how far you have come.

SECTION

Start the Conversation

Open Claude. Spend 10 minutes talking honestly about where you are in life. Then paste Prompt 1 and read the response all the way through. That first answer will tell you whether you are ready for the rest.

AI SYSTEMS ARCHITECTURE NOTE Somya's Strategic Takeaway

The real unlock of AI agents is not full autonomy, but human-in-the-loop systems design. Structure your agent inputs with explicit constraints, negative prompts, and automated test checkpoints. When building tools, keep token consumption lean by caching system prompts and isolating tasks into specialized sub-agents.

⚡ SCALING AI & GROWTH SYSTEMS?

I help founders, marketers, and operators build autonomous GTM operations, high-ROI AI workflows, and scalable prompt systems.

Book a Growth Consultation ->
Hey, I'm Mini Somya. Click me to chat.