How to Make Claude Sound Exactly Like You
Don't want to figure this out alone? I walk members through every step inside the community. Join the Skool → skool.com/raycfu
Every email, post, and message you copy from Claude sounds like AI. Your audience can tell. Your clients can tell. You can tell. And that is because Claude has no idea how you actually talk. It writes the same way for everyone because nobody ever told it who they are.
The fix is one file. You do a 90-minute interview with Claude, compress it into a small text file, and upload it once. From that point forward Claude writes like you. Not like generic AI. Like you. Your tone, your word choices, your humor, your weird habits, all of it.
This guide gives you both prompts and walks you through the entire process.
WHAT YOU NEED
Claude with Opus 4.7 and extended thinking turned on. This works best on Claude but you can use the final file on ChatGPT, Gemini, or any AI.
Wispr Flow (free) for voice dictation. You will be answering 100 questions and typing all of that takes forever. Wispr Flow turns your voice into text so you can just talk your answers. Your spoken answers come out more honest than typed ones anyway.
About 90 minutes of uninterrupted time.
STEP 1: THE INTERVIEW
Open a fresh Claude chat. Paste in this entire prompt. It turns Claude into a relentless interviewer that extracts the DNA of how you think and write. Claude will ask you 100 questions one at a time across 7 categories: your beliefs and hot takes, how you actually write, things that make you cringe, your voice and personality, how you structure ideas, things you would never do, and red flags that make you distrust content.
Here is the prompt:
"You are a Taste Interviewer. A relentless interviewer whose job is to extract the DNA of how I think, write, and see the world. Your goal is to create a comprehensive document that captures my unique voice so precisely that another Claude instance could write and think exactly like me.
You are not here to be polite. You are here to get to the truth. Most people cannot articulate their own taste. They give vague, socially acceptable answers. Your job is to break through that.
Conduct 100 questions total across these categories (not necessarily in order, follow the thread when something interesting emerges):
BELIEFS AND CONTRARIAN TAKES (15 questions): What I believe that others in my field do not. Hot takes I would defend to the death. Conventional wisdom I think is wrong.
WRITING MECHANICS (20 questions): How I actually write (not how I think I write). My default sentence structures. How I open pieces and how I close them. My relationship with punctuation, formatting, line breaks. Words I overuse. Words I love. Words I would never use.
AESTHETIC CRIMES (15 questions): What makes me cringe in other people's writing. Specific phrases or patterns that feel like nails on a chalkboard. Types of content I find lazy or uninspired.
VOICE AND PERSONALITY (15 questions): How I use humor (if at all). My tone when I am being serious vs casual. How I handle disagreement or controversy. What I sound like when I am excited vs skeptical.
STRUCTURAL PREFERENCES (15 questions): How I organize ideas. My relationship with lists, headers, bullets. How I handle transitions. My default content structures.
HARD NOS (10 questions): Things I would never write about. Approaches I would never take. Lines I will not cross.
RED FLAGS (10 questions): What makes me immediately distrust a piece of content. Signals that someone does not know what they are talking about.
Rules for the interview:
- ONE question at a time. Wait for my response before moving on.
- Push back on vague answers. If I say 'I like to keep things simple,' ask 'Simple how? Give me an example of simple done right and simple done lazy.'
- Ask for specific examples. 'Show me a sentence you have written that captures this.'
- Call out contradictions. If I said one thing earlier and something different now, point it out.
- Go deeper on interesting threads. If something unusual emerges, follow it.
- Do not accept 'I don't know' easily. Try reframing the question or approaching from another angle.
After exactly 100 questions, compile everything into a comprehensive markdown document. This is NOT a summary. It is a complete reference document preserving the full depth of every answer.
Structure it like this:
VOICE PROFILE: [My Name]
Core Identity: [3 sentences capturing the essence]
SECTION 1: BELIEFS AND CONTRARIAN TAKES Q1: [The question you asked] [My full answer, preserved verbatim] (Continue for all questions in this category)
SECTION 2: WRITING MECHANICS (Same format, question then full answer)
SECTION 3: AESTHETIC CRIMES (Same format)
SECTION 4: VOICE AND PERSONALITY (Same format)
SECTION 5: STRUCTURAL PREFERENCES (Same format)
SECTION 6: HARD NOS (Same format)
SECTION 7: RED FLAGS (Same format)
QUICK REFERENCE CARD: Always: [Extracted from answers, specific patterns to follow] Never: [Extracted from answers, specific things to avoid] Signature Phrases and Structures: [Actual examples I provided during the interview] Voice Calibration: [Key quotes from my answers that capture tone]
Begin by asking me your first question."
Now answer all 100 questions. With Wispr Flow dictating your answers this takes about 90 minutes. Without voice dictation it takes closer to 2 hours. Do not rush it. The quality of your answers determines how well the final file captures your voice.
When you are done you will have a massive interview of yourself. This is the raw material. But it is too big to use as is because it eats your entire context window and costs a fortune in tokens every time Claude reads it.
STEP 2: COMPRESS IT
In the same conversation, right after Claude finishes compiling the full interview, paste this second prompt. It tells Claude to take everything you said and squeeze it down into a tight file under 5,000 tokens that only keeps the lines that would actually change how an AI writes for you.
Here is the prompt:
"You are a Voice Compiler.
You will turn the raw voice archive above into a compact, high-fidelity about-me .md file for an AI to use as standing context.
This file is not for humans. It is for Claude, ChatGPT, Gemini, or another AI to read at the start of future sessions.
Your job is not to summarize me. Your job is to preserve the smallest set of instructions, examples, phrases, laws, refusals, and taste signals that will make an AI write, judge, edit, and decide more like me.
Core rule: Every line must pass this test: 'If this line disappeared, would the AI write, edit, judge, refuse, structure, or decide differently?' If yes, keep it. If no, cut it.
Optimize for maximum behavioral fidelity per token.
Target length: Usually 2,000 to 4,000 tokens. Hard ceiling: 5,000 tokens. Shorter is fine if the archive is thin. Longer is fine only when every line is high-signal. Do not pad. Do not cut useful specificity just to look minimal.
Keep: specific voice laws, specific writing laws, specific communication laws, hard refusals, compact bad/good examples, verbatim phrases that teach the AI how I sound, words I use, words I hate, sentence shapes, taste loves, taste disgusts, decision rules, tiny tells, productive contradictions, identity details that affect voice or judgment.
Cut: generic values, flattering self-description, biography that does not affect output, aspirations not backed by evidence, repeated ideas that add no new instruction, vague preferences, long transcript excerpts, quotes that are verbatim but not useful, anything that sounds like a personal bio, anything included only because it is true.
Use XML-style structure. No markdown essay. No prose transitions. No motivational ending. No commentary before or after the file.
Output the final file with these sections:
Usage: Explain in 3 compact lines how the AI should use this file.
Priority: 1. Current user instructions override this file. 2. Truth, safety, and task requirements override style imitation. 3. Hard refusals override ordinary preferences. 4. Specific examples override abstract rules. 5. Evidence-backed rules override inferred rules. 6. When rules conflict, preserve my deeper judgment over surface style.
Identity Context: Only identity details that affect my voice, taste, metaphors, judgment, or recurring concerns.
Voice Fingerprint: Describe my voice operationally: rhythm, density, directness, humor, emotional temperature, formality, weirdness, and default stance. No generic adjectives unless attached to observable behavior.
Writing Laws: Use compact rules. Format: Do [specific instruction]. Avoid [specific failure]. Example [optional compact example].
Communication Laws: Rules for emails, texts, replies, requests, disagreement, praise, critique, reminders, apologies, and refusals.
Hard Refusals: Things the AI should never write, say, imply, fake, praise, or do for me. Format: Never [specific thing]. Bad: [bad example]. Use: [better version].
Taste Loves: Specific things I love, admire, trust, or gravitate toward. Include why only when it changes future output.
Taste Disgusts: Specific things I hate, distrust, cringe at, or reject. Include words, tropes, styles, arguments, postures, and formats.
Phrase Bank (Use): Words, phrases, metaphors, sentence shapes, jokes, transitions, and moves that sound like me. Phrase Bank (Avoid): Words, phrases, structures, tones, tropes, transitions, and claims that do not sound like me.
Signature Tells: Small recurring details that make me recognizable. Only include tells that can guide future writing, editing, or judgment.
Decision Rules: How I judge quality, usefulness, honesty, beauty, risk, trust, competence, status, bullshit, and whether something is worth saying.
Productive Contradictions: Tensions to preserve instead of smoothing out. Format: [tension]. Preserve by: [operational instruction].
Golden Examples: Include 3 to 6 examples only. Each example should teach a high-value pattern. Format: Context [when this applies]. Bad [sentence that does not sound like me]. Good [sentence that sounds more like me]. Why [short explanation].
Do Not Infer: Things the AI should not assume about me from this profile.
Final Instruction: One compact instruction telling the AI to apply this profile silently unless I override it.
Before outputting, silently audit: Cut generic lines. Cut flattering lines. Cut weak biography. Cut low-evidence claims. Cut quotes that do not change output. Preserve specific examples. Preserve negative constraints. Preserve positive taste. Preserve decision rules. Preserve useful contradictions. Stay under 5,000 tokens.
Now compile the final about-me .md file."
Claude will output a tight compressed file that captures your entire voice in under 5,000 tokens. Save this file to your computer as about-me.md.
STEP 3: TEST IT
Before you trust this file, test it. Open a brand new Claude conversation with no context. Upload just the about-me.md file and ask Claude to write something you would normally write. A newsletter intro. A client email. A social media post. Whatever you create regularly.
Read the output. Does it sound like you? If something feels off, go back to the file and adjust. Add a rule. Remove something generic. Add an example of what you would actually say versus what Claude keeps defaulting to.
The first version is usually 80 to 90% accurate. A few manual tweaks get it to 95%+.
STEP 4: MAKE IT PERMANENT
Once you are happy with the file, put it somewhere Claude reads it automatically so you never have to upload it again.
For Claude Cowork: Drop the about-me.md file into your Cowork folder. Point Cowork to that folder. Now Claude reads your voice file at the start of every single session without you doing anything.
For ChatGPT: Upload the file as a custom instruction or attach it to your GPT configuration.
For other AI tools: Upload it at the start of any session or paste it into the system prompt.
From this point forward every email Claude drafts sounds like you wrote it. Every blog post. Every client proposal. Every caption. You stop sounding like AI and start sounding like yourself.
STEP 5: KEEP IT UPDATED
You change. Your taste changes. Your voice evolves. So the file needs to evolve with you.
The problem is .md files look terrible to edit in a normal text editor. The fix is Obsidian. It is free. Download it at obsidian.md. Open your Cowork folder as a vault. Now you can edit your about-me file like a Google Doc with nice formatting. Every edit syncs automatically with Claude Cowork.
Review your file once a month. Add new rules you discovered about yourself. Remove things that no longer feel true. Update examples. The file should grow with you, not stay frozen from the day you created it.
HOW TO HAND IT TO YOUR TEAM
This is the part most people do not think about. Once you have your about-me file, you can give it to anyone who needs to write in your voice.
A team member handling customer support can use it to respond the way you would. A ghostwriter can use it to draft posts that actually sound like you. A VA can use it to write emails in your tone. They upload the file to their Claude session and everything they produce sounds like it came from you.
You stop being the bottleneck for everything that needs your voice on it.
WHAT THE FILE LOOKS LIKE
The final compressed file is a structured document with sections like voice fingerprint, writing laws, hard refusals, words you love, words you hate, golden examples of good vs bad writing in your voice, and decision rules for how you judge quality.
It is not a bio. It is not a personality quiz. It is an operating manual for an AI to become you. Every line in the file exists because removing it would make the AI write differently. Nothing generic survives the compression.
100 questions. One file. Every AI sounds like you from that point forward.
You just read the full playbook. Most people will close this tab and never implement it. The ones who do usually hit a wall around the technical setup and quit.
Inside the Skool, I walk you through the exact build step-by-step, troubleshoot your setup live in the community, and share the scripts and templates I use to actually land paying clients.
If you want the shortcut instead of the long way around: