This is the week you launch your Jarvis.

What it does

You paste this template into a Skill and answer it by voice, two or three sentences per line. From then on, any chat where you ask about your business already knows your market, your niche, where your business actually comes from, and what you're building toward. You stop re-explaining yourself, forever.

Why it's worth it

Ask a blank chat what to focus on this week and you get an answer that could be for anyone, in any city, at any price point. That's not the AI being dumb; about your business, it's a stranger. One sitting of talking fixes that permanently, and every use case further down this page gets sharper the moment it exists.

A few fields in here are worth handling carefully: revenue, deal size, what your competitors do well. Give a range, describe a pattern, skip the names. Full version of that rule is in "The rules I follow" at the bottom of this page.

The template

Eight sections, a long list of fields, and nobody finishes it in one sitting. Fill in what you can this morning. The rest is tonight, in the car, by voice, the same way you did your Personal DNA.

The first two lines are new — a name and a description. Don't skip them. That description is what makes this a Skill instead of a document: Claude reads it before every single message you send, and the moment your question is about your business, it pulls this whole file in on its own. You never have to remember to open anything.

---
name: business-dna
description: Reference this whenever I ask about my business — my market, my niche, my clients, my leads, my goals, my revenue, my competitors, my team, or my marketing. Use it to make every answer specific to me instead of generic, and if something below is missing, ask me instead of guessing.
---

# MY BUSINESS DNA

## Core Business Identity
- Business Name:
- Mission Statement: (What is your ultimate purpose beyond making money?)
- Vision Statement: (What does success look like in 2-3 years?)
- Core Values: (The 3-5 principles that guide every decision)

## Products/Services & Value Proposition
- Main Products/Services: (List and briefly describe each)
- Unique Selling Proposition: (What makes you different from competitors?)
- Core Benefits: (What problems do you solve? What transformation do you deliver?)
- Pricing: (General pricing structure — ranges are fine)

## Target Market
- Ideal Client Profile: (Who exactly do you serve? Be specific: age, role, industry, situation)
- Where They Hang Out: (LinkedIn? Instagram? Industry events? Specific communities?)
- What They Say When They're Frustrated: (Their exact words, not your marketing language)
- What They've Tried Before: (What solutions have they attempted that didn't work?)

## Business Goals & Challenges
- Primary Goal (next 90 days):
- Secondary Goals:
- Biggest Challenge Right Now:
- What's Holding You Back:

## Revenue & Business Model
- How You Make Money: (Retainers? Projects? Products? Subscriptions?)
- Average Deal Size:
- Current Monthly Revenue Range: (Ballpark is fine)
- Revenue Goal (12 months):

## Competition & Positioning
- Top 3 Competitors: (Names + what they do well)
- What You Do Better:
- What They Do Better: (Be honest — your Clone needs the full picture)

## Team & Resources
- Team Size:
- Key Roles: (Who does what?)
- Tools You Use Daily: (CRM, email, project management, etc.)
- What You Wish You Could Delegate:

## Content & Marketing
- Platforms You're Active On:
- Content That Works Best: (What gets engagement? What converts?)
- Content You Struggle With:
- Brand Voice: (Point this at the voice profile you built in Week 2)

A few fields here are worth handling carefully: revenue, deal size, what competitors do well. Give a range, describe a pattern, skip the names.

How to set it up

Why this is a Skill and not a project

Your Personal DNA from Week 1 goes in your account settings, so it follows you into every chat you ever open, no matter what you're talking about. Your Business DNA is different: it should only show up when you're actually asking about your business. A Skill does exactly that. It sits quietly until your question matches its description, then it loads itself. Same as your Personal DNA in one way — you never have to remember to turn it on — and better in another: it doesn't clutter every unrelated chat. Works the same whether you're in a project, a plain chat, or Claude Code. That's the whole idea behind a Skill: it's not a place you go, it's a description that finds you.

What it does

You just had a conversation with AI that worked — you went back and forth, gave feedback, rejected a few drafts, and landed on something good. Most people walk away with the output and lose the process. This prompt reads the whole conversation and pulls the method out of it: the steps you actually followed, the things you kept correcting, the quality bar you were applying without saying so out loud. Then it writes that method up as a playbook you can hand back to AI and run again in one shot.

Why it's worth it

A twenty-minute back-and-forth becomes a one-time setup. The next time you need the same kind of thing — a comp analysis, a follow-up email, a listing description — you're not starting from a blank page and re-explaining your taste. You're pasting in a playbook that already knows how you like it done.

The prompt

You are a Framework Extraction Specialist. Your job is to reverse-engineer implicit methodologies from AI chat conversations.

I'm going to give you a complete conversation between a human and an AI. The human probably thinks they were "just chatting" — iterating, refining, going back and forth until they got a result they liked.

But hidden inside that conversation is a METHOD. A repeatable framework. Steps they followed — maybe without realizing it. Patterns in how they gave feedback, what they prioritized, how they refined, what they rejected, and what made the final result work.

Your job: find that method. Extract it. Name it. Document it as a reusable playbook.

## ANALYSIS PROCESS

Phase 1 — Read & Map
Read the entire conversation. Map every meaningful exchange:
- What did the human ask for initially?
- What context did they provide (and what was missing)?
- Where did they redirect, correct, or refine?
- What quality criteria did they apply (explicitly or implicitly)?
- What was the final output that satisfied them?
- What made the final version different from the first attempt?

Phase 2 — Pattern Extraction
Identify the underlying framework:
- What SEQUENCE of steps did they actually follow?
- What DECISIONS did they make at each step (and what informed those decisions)?
- What QUALITY FILTERS did they apply? (what did they reject and why?)
- What INPUTS were essential vs. optional?
- What DOMAIN KNOWLEDGE did they bring that the AI didn't have?
- Were there any PIVOT POINTS where the direction changed significantly?

Phase 3 — Method Documentation
Turn your analysis into a clean, reusable playbook.

## OUTPUT FORMAT

Produce a single document in EXACTLY this structure:

---

# [GIVE THE METHOD A NAME]
[One-sentence description of what this method does]

## Origin
Extracted from a conversation where [brief description of what was built/created].

## What This Method Does
[2-3 sentences: what problem does this solve, and what does the output look like?]

## When To Use This
- [Trigger situation 1]
- [Trigger situation 2]
- [Trigger situation 3]

## Required Inputs
Before starting, you need:
1. [Input 1 — what it is and why it matters]
2. [Input 2]
3. [etc.]

## The Framework

### Step 1: [Name]
What to do: [Clear instruction]
Why this matters: [What goes wrong if you skip this]
Quality check: [How to know this step is done well]

### Step 2: [Name]
What to do: [Clear instruction]
Why this matters: [What goes wrong if you skip this]
Quality check: [How to know this step is done well]

[Continue for all steps...]

### Step N: [Name]
What to do: [Clear instruction]
Why this matters: [What goes wrong if you skip this]
Quality check: [How to know this step is done well]

## Quality Filters
Things to always check / reject / refine:
- [Filter 1 — from what the human rejected or corrected during the conversation]
- [Filter 2]
- [Filter 3]

## Anti-Patterns
Things that went wrong during the conversation that this method now avoids:
- [What didn't work and why]

## Claude Skill Instructions
Copy everything below this line into Claude as project instructions or a skill: You are an AI assistant that follows the [METHOD NAME] framework.  When the user provides [REQUIRED INPUTS], execute the following steps in order:  [Compressed version of the framework above, written as direct AI instructions]  After completing all steps, present the result and ask: "Want me to refine any step, or is this ready?"
---

## IMPORTANT RULES

1. Extract what's ACTUALLY there. Don't invent steps that weren't in the conversation. If the human only followed 3 steps, document 3 steps. Don't pad it to look more impressive.

2. Name the method something memorable. Not generic ("Content Creation Process") but specific and evocative ("The Objection-First Proposal Builder" or "The Voice-to-Authority Pipeline").

3. Capture the human's taste. The most valuable part is often WHY they rejected certain outputs. That's their quality filter — document it explicitly.

4. Be honest about what you find. If the conversation doesn't contain a clear reusable framework, say so. Not every chat has a hidden method. Output: "No extractable framework found. This conversation was [exploratory / one-off / too short to identify a pattern]. Here's what I did find: [any partial insights]."

5. The Claude Skill section must be copy-paste ready. No placeholders the user needs to fill in. It should work immediately when pasted into Claude.

---

Now analyze this whole chat from the beginning.

How to run it

What they are

Everything I demo at this bootcamp runs on systems I actually use. These two are written down completely: what the system does, why the design matters, the exact prompts, and the compliance lanes to stay inside. Each one works two ways. Use my version and I'll set you up, or hand the doc to Claude Code and build your own.

The shelf

What it is

The write-up behind the story from today: 584 lessons read overnight, comps pulled off a real estate site live in front of a room, campaigns built in software I never learned, all while I was at the mall with my girls. Six real jobs, the exact prompt for each one, and the rules I'd give you before you start.

Why it's worth it

Most people ask AI for an answer and stop. The jobs in here are different: you give it a browser and a task, and it goes and does the thing instead of describing how you'd do it. That one change is worth more than any single prompt.

How to get it

What it is

This is the Content to Clients arc from today, written down step by step: what to post and why, the five daily opens, the three-line DM opener, the close, and the gift that turns a follower into an actual conversation. The frameworks are Dan Martell's; the adaptation to real estate is mine.

Why it's worth it

Posting without a system to turn attention into conversation is wasted effort. This is the system. Not more content ideas — the part that comes after you post, which is the part almost everyone skips.

How to get it

What it does

One prompt, run in any chat, that generates ten ways to use AI this month that are specific to your market, your niche, your lead sources, and your goals. Then you hit Retry and get ten more. Twenty tailored ideas in five minutes.

Why it's worth it

This list doesn't exist on any blog, because it's generated from your Business DNA. And here's the part that matters: nothing about the AI changed. What changed is what it knows about you. That's the whole point of the work you just did.

The prompt

You know my business from my Business DNA. Generate 10 hyper-relevant, immediately actionable ways I can use AI in my business this month. For each one: what it is in one sentence, why it makes me money or saves me time, and the exact first step to take today. Rank them by payoff for effort. Nothing generic. Every idea must be specific to my market, my niche, my lead sources, and my goals.

How to run it

What it does

You type one line with an address in it. What comes back is the comp set, a full written analysis with a value range and the reasoning behind it, a designed flyer in your branding, and the email to the client already drafted in your voice. Two versions of the last two, actually: the seller version ends in a pricing strategy, the buyer version ends in an offer strategy. Same data, two audiences, one prompt each.

Why one line is enough

Because everything the AI needs to know is already sitting in your project. Your branding, your format, how you write, who the client is. You built that this morning. This is the whole Week 4 lesson in one artifact: the short prompt only works because the context is already there. In a blank chat the same sentence gets you a Wikipedia article about comparative market analysis.

Step one, the line you actually type

Do a CMA for [address]. Pull the comps, analyze them, and build me the listing presentation and the email.

Step two, the routine that makes it good

This is the part that does the work. You add it to your Business DNA Skill once, and from then on the one-liner above triggers all of it. Read it before you use it. Most of what is in here is the difference between an analysis you can defend to a seller and a pretty document that is wrong.

When I ask you for a CMA, run this whole routine and don't stop to check in unless something is genuinely ambiguous.

STEP 1 - THE SUBJECT
Pull everything available on the subject property: beds, baths, gross living area, lot size, year built, stories, garage, pool, HOA, tract or subdivision, school attendance area, and last sale date and price. Tell me what you could not find; do not guess and do not fill a gap with a neighborhood average.

STEP 2 - CAST THE NET
Search closed sales in the last 6 months. Start inside the subject's own tract or subdivision. If that gives fewer than 6, widen to a half mile, then one mile, then the school attendance area. Stop widening the moment you have 6 to 10 good ones.

Never cross a hard market boundary just to get closer in distance. A freeway, a major arterial, a city line, or a school district line separates two markets even when the houses are 800 feet apart. Say out loud which boundaries you respected.

Similarity bands, tightest first:
- Gross living area within 10%, and never past 20%
- Same bedroom count; baths within one
- Year built within 10 to 15 years
- Lot within 25%
- Same story count and same general style

STEP 3 - PULL THE THREE OTHER DATA SETS MOST PEOPLE SKIP
- ACTIVE listings in the band. These are my competition, and they set the ceiling.
- PENDING sales. These are the most current read on direction, ahead of closings.
- EXPIRED and WITHDRAWN in the last 12 months. This is the sharpest evidence I can get about where the market said no. Tell me what they were asking.

STEP 4 - ADJUST, CORRECTLY
Adjust each comp TO the subject. Comp is better than the subject, subtract from the comp. Comp is worse, add to it.

Use contributory value, not full price per square foot, for size differences. Full price per square foot double-counts the land and is the single most common error in agent CMAs. Where you can, derive the adjustment from paired sales in this same data set and show me the pair you used.

Apply a market-conditions (time) adjustment based on how the local median price per square foot has actually moved across the comp window. Do not skip this and do not assume zero.

Check every comp against the adjustment guardrails: net adjustment within 15% of sale price, gross within 25%, no single line item over 10%. Any comp that blows past those is not comparable. Drop it, and tell me you dropped it and why.

Read the concessions. A sale with seller credits is not the number on the line.

STEP 5 - RECONCILE, DON'T AVERAGE
Do not average the adjusted values. Weight them: the comps needing the fewest and smallest adjustments carry the most weight. Give me:
- An indicated value range, tight
- The 3 comps you leaned on hardest, and why those three
- What would move the number up or down, and by how much

STEP 6 - THE STRATEGY LAYER
- Months of supply and absorption rate in this specific price band, not the market headline
- Median days on market, and list-to-sale ratio in the band
- The price-band break points. Buyers filter in round numbers. Tell me what listing at $1,025,000 costs me versus $999,000.

STEP 7 - TELL ME WHAT YOU DON'T KNOW
List every assumption you made and every input you're missing. Condition, updates, views, noise, and the interior are things you cannot see. Ask me for them.

STEP 8 - THE DELIVERABLES
Build, using my branding and my format from my Business DNA:
- The seller version: listing presentation, comps laid out so the comparison is obvious at a glance, a recommended pricing strategy, ready as a PDF
- The buyer version: the same data, ending in an offer strategy instead of a list price
- An email to each, in my voice

Never put a valuation conclusion in a document as if it were an appraisal. Give me the range and the reasoning. The number I say out loud to a client is mine, not yours.

The three places agent CMAs actually break

Where you still matter, and it is not close

In that same data set, one house sold twice inside the study window. $935,000 in March, $1,182,000 in June. Same square footage, same lot, same year built. $247,000 in 118 days, and all of it was condition, meaning what somebody did to the inside. There is no MLS field for that. The machine cannot see it and never will, and reading it is the part of this job that pays.

The rule, and it is not optional

The value conclusion is yours. It is never the AI's. This produces a range and the reasoning behind it, which is exactly what it should produce. Nothing it hands you goes to a client as an appraisal or as a fixed number, and nothing goes out with a value on it that you have not personally stood behind. You hold the license. Read every comp it picked before the document leaves your hands.

How to set it up

What it does

Instead of one task in one chat while you sit there watching it, you write down everything you want done, hand the whole list over at the end of the day, and go to bed. It works through the stack while you sleep. In the morning the drafts, the research and the files are sitting there finished, and your job is to read and approve instead of to produce.

Why this one is the headline

I want to be honest about how long this took me to get. I heard people say "have your agent work overnight" for months and I nodded along and had no idea what they actually meant. It clicked last weekend when I watched a chain run end to end without me: pull the data, build the thing, load it into the tool, and have it all waiting at 6am.

Here is the shift, and it is the only thing on this page I'd ask you to actually remember: an agent is not a faster chat window. It's a night shift. You stop thinking "what can I ask it right now" and start thinking "what should be done before I wake up."

Your version

The prompt that starts it

I'm going to hand you a stack of work to run tonight while I sleep. Before you start anything, do this:

1. Read my list back to me as a numbered plan, in the order you'll actually do it, and tell me roughly how long each piece takes.
2. Tell me which items you CAN'T finish without me, and what you'd need from me to unblock them.
3. Tell me which items you think are a bad use of the night, and why.

Rules for the whole run:
- Draft everything. Send nothing, post nothing, schedule nothing, and don't message anyone.
- If you hit something ambiguous, make your best call, DO IT, and write the assumption down instead of stopping.
- Write one summary file I can read in under 5 minutes in the morning: what got done, what didn't, what needs a decision from me, and where every file is.

Here's the stack:
[talk it out - say everything you want done tonight, in any order]

How it works

This is the Claude Code lane, on your laptop, and it's the reason we install it. Two practical things I learned the hard way. Tell it to make a call and keep going instead of stopping to ask, or you'll wake up to a question and no work done. And make it write the summary file. A night of work you don't have time to read is a night of work that didn't happen.

One more honest note, because it's the real risk: the point of the night shift is to buy back your evenings, not to fund later ones. If you end up staying up until 1am loading the queue, you've traded the wrong thing.

What it does

You point it at a training library you have access to. Overnight it walks the whole thing, pulls the transcript of every lesson, downloads the files and templates, and sorts all of it into three piles: must watch, skim, and skip. Not by popularity. By whether it touches something you are actually working on right now.

Why it's worth it

I paid for a membership, and a month later I had not touched it. Not because it was bad. Because there were 584 lessons in there, roughly a hundred hours of video, and I had no way to know which six hours were the ones for me. So I set it running and went to bed. Saturday at the pool while my girls were swimming, I read the whole summary in an hour, then watched only what mattered.

Then the part I did not expect. It had also found 1,336 downloadable files in there, which is the half of a membership that keeps its value after you cancel. Videos go stale. Files do not.

Your version is sitting in your account right now. Mike Ferry, Tom Ferry, Ninja, your brokerage's university, whatever you bought in January and opened twice. You do not have to watch four hundred hours to know which six matter.

The prompt

I pay for [name of the program] and I have gotten through almost none of it. I want to know which parts are worth my time and which I can skip forever.

Here is what I am actually working on right now: [3-5 real things. "Get more listings in [farm]" / "my sphere has gone cold" / "I have no follow-up system" / "I want to start posting"].

Go through the library lesson by lesson. For each one, pull the transcript or the description, then sort it into exactly one of three piles:

MUST WATCH - it directly addresses one of the things I listed. Tell me which one, and what specifically I would get out of it.
SKIM - there is one useful idea in it. Give me the idea in two sentences so I do not have to watch it at all.
SKIP - say why in half a line. Out of date, not my market, or I already do it.

Rules:
- Rank the MUST WATCH pile in the order I should actually watch them, and give me the total runtime so I know what I am committing to.
- Grab every downloadable file, script, template and checklist. Those keep their value after the videos go stale. Tell me which ones map to something I am doing.
- Do not be generous. If most of it is skip, say most of it is skip.
- Write me one summary I can read in five minutes.

How it works

The full overnight version is the Claude Code lane, because it needs to sit there for hours working through a library while you sleep. That is next week.

Your version today, no install: take one course, not the whole library. Copy the lesson list off the course page, paste it into Claude with the prompt above, and let it triage the titles against what you are working on. You will not get transcripts, but you will get a watch order, and a watch order is the thing that was actually missing.

Two lanes to stay inside. Most memberships allow personal-use archiving and forbid bulk export, so read the terms of the one you pay for before you point anything at it, and throttle it rather than hammering it. And this is for a library you currently pay for, on your own login. It is not a way to get at one you don't.

What it does

Every day your account collects signals: new followers, likes on your last post, and the list of people who watched your story. AI reads each of those profiles and hands you a short row per person: who they are, whether they're an agent, a possible buyer, a possible seller or not in real estate at all, what in their profile you could genuinely reference, and a first message written for that one person.

Then you open Instagram and send five. By hand. From your phone. Nothing on this card ever sends anything for you. It removes the part where you sit there wondering who to message and what to open with. That was always the whole job: the blank page, not the typing.

Why it's worth it

This is the warmest audience you will ever have and almost nobody works it. Someone who watched all the way through your story today is warmer than any lead you can buy, and they cost you nothing. The bottleneck was never sending the message. It was knowing what to say to a stranger's profile, and that is the entire thing this solves.

One finding from my own runs worth stealing: on the nights when new followers and post likes were both dry, the story viewer list was the vein that still produced. Most agents never look at it. It's right there under the story, next to the number.

Start here · no agent, five minutes, tomorrow morning

You do not need anything built to do this. Open your story, tap the viewer list, screenshot it, and hand the screenshot to Claude with the prompt below. That is the whole setup. Do it once and you have done the five opens for the day.

Here are the people who watched my story. I'm a real estate agent in [your market] and I work mostly with [who you work with].

Write me one text for each of the first five, warm, two sentences max, in my voice. Each one has to be about THEM, not about me. No pitch, no "let me know if you're ever thinking of selling," no link.

If you can't tell anything real about someone from what I gave you, say so and skip them instead of writing something generic.

I'm sending these myself, one at a time. Draft only.

The full version, once you want the research done for you

I'm going to give you a list of Instagram handles - people who just followed me, liked my last post, or watched my story. I'm a [realtor / lender] in [your market].

For each person, look at their public profile and give me one row:

1. Name and handle
2. What they are: real estate agent / likely buyer / likely seller or homeowner / other industry / not relevant
3. One specific, real detail from their profile I could genuinely reference. Not "loves travel." Something specific.
4. A first message, 2 sentences max, in my voice. It must be about THEM. No pitch, no "let me know if you ever need a lender," no link.
5. Skip flag: mark anyone who looks like a bot, a business page, a minor, or someone clearly outside my market.

Rank them by who is most worth my time today, and give me the top 5.

IMPORTANT: draft only. Do not send, like, follow or comment on anything. I send every message myself.

Here's the list:
[paste handles]

The rules that keep your account alive

What it does

The card above finds you five people to talk to. This one keeps the conversation alive after they answer, which is where almost everybody loses the deal. It tracks every person you've messaged, whose turn it is, how many days it's been, and what they last said — then hands you the next message already written.

You stop opening a chat, scrolling up to remember what happened, doing the date math in your head, and then staring at the box.

Why it's worth it

The single most valuable thing it tracks is who spoke last. If someone replied to you and you haven't answered, that is an emergency — a warm person going cold because you left them on read is the worst thing that happens in this whole system, and it happens constantly. That one column belongs at the top of your list every morning.

The honest scoreboard from my own account: I had thirty drafted openers and had sent almost none of them. Research was never the bottleneck. The follow-up half is what turns five opens a day into a pipeline instead of a chore.

Step 1 · set it up once

Make a Claude Project and paste this into the project instructions. One time, then you never touch it again.

You are my daily outreach engine. I'm a real estate agent.

ABOUT ME
- Market: [your city / the areas you actually work]
- I work mostly with: [first time buyers / move-up sellers / investors / your sphere]
- My voice: [warm and casual / direct / funny] - short sentences, no corporate language
- What makes me different: [the real answer, one sentence]

MY GOAL
Five NEW conversations a day, plus keeping the existing ones alive.
The goal of a first message is a reply. Nothing else.

THE RULES - these never change
1. DRAFT ONLY. You never send, like, follow or comment. I send everything myself.
2. Every message is about THEM, not me. If I can't say something specific and
   true about the person, skip them.
3. No pitch in the first message. No link in the first message. Ever.
4. Never ask if they're thinking of buying or selling in message one.
5. Never guess at or comment on anyone's race, religion, family status,
   nationality, or neighborhood "type." Fair housing line, absolute.
6. Skip anyone who looks like a bot, a business page, or a minor.
7. If a conversation has run its course, tell me to let it go. Do not invent a
   reason to keep poking someone.

FILES YOU KEEP FOR ME
- contacts.csv - everyone I've ever talked to, one row each
- today.md - today's five new + today's follow-ups

Step 2 · the one you run every morning

Here's today's batch. [paste screenshots of your notifications and story viewers]

Do three things:

A) FOLLOW-UPS FIRST. Look at contacts.csv. Tell me:
   - Who replied and I haven't answered yet   <- always the top of the list
   - Who I messaged 7+ days ago with no reply  <- one nudge, once, ever
   - Who's had a real conversation and is ready for an actual ask
   For each, remind me what they last said and draft my next message.

B) THE NEW FIVE. From the batch above, rank everyone and give me the top five:
   name, what they are, the one specific detail, and my first message.

C) UPDATE contacts.csv with everyone new, and mark anyone I told you I sent.

Keep the whole thing under 10 people. If there are more, show me the best ten
and tell me how many you held back.

What to actually say · the ladder

Don't skip rungs. Most agents jump from hello straight to "let me know if you need anything real estate," which is exactly why it doesn't work.

1. Open · one specific thing about them  →  2. Reward · acknowledge, then ask again, same message  →  3. Their world  →  4. What they want  →  5. What's in the way  →  6. Dig  →  7. Bridge · "that's what I do all day," one line, then stop  →  8. The ask

Rung 2 is the one everybody skips. They answer and you either go quiet or you pitch. "Nice, beautiful area." ends the conversation. "Nice, beautiful area. How long have you been out there?" keeps it. Acknowledge, then ask, in the same message.

And the rule that keeps you from being annoying: no problem, no next step. If you get to the bottom and there's nothing they actually need, you have a friend, not a client. That's a fine outcome. Stay in touch and stop selling.

The follow-up clock

The full written playbook

Open the complete 5 Opens playbook — the five-minute starter version, the full system, the tracker columns, the question bank, and the follow-up clock in one doc you can work from.

What it does

You start with one real thing that happened: a rule change, a rate move, a listing in your farm that sold way over. You hand it to AI once and ask for the whole chain. Back comes a text short enough to actually send, an email for your database, a carousel with the slides written, and the caption for the post. Four pieces, one source, one sitting.

Why it's worth it

Two people in this room have told me they do not have time for content. They are right, if content means coming up with something new five times a week. That is the part that is impossible, and it is also the part nobody actually has to do.

This morning a brief I did not write landed at 6am with a rule change in it. I did not come up with the idea and I did not write the first draft of anything. By 9 the text list, the email, the carousel and the post were done. The idea was free. The writing was free. All I did was decide it was worth saying and put my name on it.

The prompt

Here is one thing that just changed in my market: [paste the article, the rate sheet line, the MLS listing, or just say it in a sentence].

I am a real estate agent in [market]. My people are [who they are].

Turn this one thing into four pieces. Same message, four rooms:

1. A TEXT to my sphere. Under 300 characters, sounds like a person, no link unless it genuinely helps. It should read like I thought of them, not like a broadcast.
2. An EMAIL to my database. Subject line plus body. Lead with what it means for them, not with what happened. Short paragraphs, one idea each.
3. A CAROUSEL, 5 slides. Slide 1 is the hook and it has to be readable in two seconds. Slides 2 through 4 each carry one idea. Slide 5 tells them what to do next.
4. An INSTAGRAM CAPTION for that carousel, in my voice, ending with a line that tells people exactly who to send it to.

Rules:
- Every claim has to trace back to what I gave you. If you are not sure of a number, leave it out. I would rather say less than say something wrong.
- Answer "so what does that mean for me" in the first two lines of every single piece. Nobody cares what changed, they care what it costs them.
- No hype, no exclamation points, no emojis.
- Then tell me the one piece you would actually send if I only had time for one.

How to run it

What it does

You describe the one thing people would trade an email address for. Claude builds you a one-page capture form as a single file: headline, one or two fields, a button, and it works on a phone. You paste the code into your website, and from then on every person who fills it out lands somewhere you can reach them again.

Why it's worth it

You can post every day for a year and still own nothing. The follower belongs to the platform. The email address belongs to you, and it is the only piece of an audience that survives an algorithm change, a hacked account, or a platform deciding your posts are worth less this quarter.

This is also the missing half of the card above. Turning one market change into four pieces is good. Turning it into four pieces that all point at one page where someone can raise their hand is the difference between reach and a pipeline.

The prompt

Build me a one-page email capture, as a SINGLE self-contained HTML file I can paste into my website. No frameworks, no external files, everything in the one file.

What people are signing up for: [the guide / the neighborhood report / the open house list / the monthly market note].
Who they are: [first-time buyers in X / homeowners in my farm / whoever].
My branding: [name], [brokerage], [DRE #], [phone], colors [x and y].

Build it with:
- A headline that says what they get, not what I do.
- Two fields, first name and email. Nothing else. Every extra field costs me signups.
- One button, and the button says what happens next, not "submit."
- A one-line note under the form telling them what I will and will not send.
- Reads well on a phone first. Most of my traffic is a thumb.
- Send the form data to [where it goes - my CRM form endpoint, or a Google Sheet, or tell me my options].

Then tell me, in plain language, exactly where to paste this on my site and what to test before I share the link.

Where the names actually go

The part people skip

Have something to give. A page that says "sign up for my newsletter" collects almost nobody. A page that says "the one-page guide to what the new rule means for your HOA" collects people, because it is worth the trade. Build the thing first, then build the page for it. The card above is how you build the thing.

If you are going to text anyone who fills this out, say so on the form, in writing, right where they type the number. Consent has to be collected at the moment they hand you the contact information, and an email opt-in is not permission to text.

What it does

You record yourself talking — in the car, at your desk, one take, flubs and all. AI listens to the recording, finds the takes worth keeping, and cuts the rest. Every cut gets masked with a small zoom change so it reads as an intentional edit instead of a jump. Then it burns in the captions, synced word by word to your voice, puts a written hook over the opening seconds, and closes with your call to action.

Why this one matters

The talking-head video is the one everybody agrees they should be posting and almost nobody posts. It's not the recording that stops you — it's the hour of editing after. This removes the hour. You do the one thing AI can't: be you on camera. It does everything after.

Where to start

Record one take on your phone this week — 60 to 90 seconds on the question buyers keep asking you. Don't restart when you flub, just say the line again and keep going. That's the raw material the edit is built from, and it's the version I'll walk through live.

This is the deeper lane — the edit runs in Claude Code, the same lane as Mission Control below. If you're not there yet, the Reel card on the AI tools page gets you most of the way with just the app: cut plan, captions, cover, and post copy from one conversation.

What it does

AI takes the wheel of a browser tab you're already logged into and does the clicking. Not a new app, not an integration, not a subscription. The same screens you use, operated for you. Mine pulls MLS searches and exports them, and loads video emails into BombBomb, which is a job I had been doing by hand every single week.

Why it's worth it

Almost every tool an agent pays for has no way to talk to anything else. The MLS doesn't hand your data to your CRM, your CRM doesn't hand it to your email tool, and the glue in between has always been you, at a keyboard, at 10pm. This is the first thing I've used that actually fills that gap, because it works the way you work: through the screen.

Where it earns its keep

The honest caveats

It breaks when a page changes. A vendor ships a redesign and your setup needs a nudge. It is slower than you, step for step, which stops mattering the moment it's running at 2am. And it is logged in as you, so treat it exactly like handing your laptop to an assistant: watch the first few runs, and never point it at anything that sends, pays or signs.

Check your MLS rules before you point it at the MLS. Data licensing and scraping terms are real, they vary by board, and "the AI did it" is not a defense. Pulling your own permitted searches on your own login is a different thing from harvesting the database, and you need to know which one you're doing.

What it does

You give AI something you pay for or something you run, and instead of asking "is this good," you ask it to make the case for cutting it and show the evidence either way. You get back what you're actually using, what you're actually getting, what's genuinely valuable in there, and a keep-or-cut verdict with a date on it.

Why it's worth it

Two audits I ran in the last week changed real decisions. One was a training membership I'd been paying for monthly. The audit found I had gotten through about 2% of it, that finishing it was roughly a hundred hours of video, and that the real value wasn't the videos at all but a library of downloadable files I could keep forever. That flipped the decision from "keep paying and try harder" to "take the files this weekend, then cancel." I would not have found that by feeling guilty about it for another month.

The second was an audit of my own automations, and the finding was that I had built roughly three times more of them than I actually had running. The fix wasn't building more. It was cutting.

This is the use case nobody teaches, because every AI demo is about making more. The highest-leverage question you can ask is what to stop.

The prompt

I want an honest audit of [the thing - a subscription, a CRM, a lead source, a marketing spend, a system I run]. It costs me [$X/month] and [roughly how much time per week]. It renews on [date].

Do not be encouraging. Your job is to build the case for CUTTING it, and then tell me honestly whether that case holds.

Give me:
1. What I'm actually getting for the money - the specific, provable value, not the sales page.
2. What I'm actually using. Be blunt about the gap between what I bought and what I touch.
3. What's genuinely valuable in here that I'd lose, and whether I could keep it another way.
4. What it's costing me beyond the money - my time, my attention, the thing I'm not doing instead.
5. The verdict: keep, cut, or downgrade. One recommendation, not three options.
6. If the verdict is cut: exactly what I should take out of it first, and by when.

If you don't have enough information to answer any of these, tell me what to go look up. Don't guess and don't pad.

Your version

Run this on your CRM, your paid portal leads, your lead-gen subscription, your photographer, your ISA, your farm mailer. Anything with a monthly number on it. Do it two weeks before the renewal date, not two days after. The only reason a bad subscription survives is that the renewal is always a surprise.

The one line that makes it work is "do not be encouraging." Without it you'll get a balanced summary that lets you keep everything.

What it does

You sit down once and dump every unfinished thing out loud: the escrow you're worried about, the person you owe a call, the thing your broker asked for, the paperwork, the thing you keep meaning to fix on your website. AI sorts it into what's actually urgent, what's a two-minute task, what needs a decision from you, and what you should just kill. Then it asks you the questions you've been avoiding.

Why it's worth it

I did one of these last weekend and it took most of a morning. It surfaced somewhere around sixty open loops, and a real number of them were things I had genuinely forgotten existed. That's the part worth noticing. It isn't that I was behind on sixty things. It's that some of them had been quietly rotting because nothing in my week was ever going to surface them.

For an agent the version of this is simple: your head is not a database, and the loops it drops are the expensive ones. The client who never got the follow-up. The referral you never thanked anyone for.

The prompt

I'm going to talk for ten minutes and dump every open loop in my business. Deals, people, paperwork, things I said I'd do, things I've been avoiding. It'll be out of order and I'll repeat myself. Don't interrupt and don't clean up my language.

When I'm done, give me:
1. NEEDS ME TODAY - anything with a deadline or a person waiting on me. Ranked.
2. TWO MINUTES EACH - the small things I can clear in one sitting.
3. NEEDS A DECISION - things that aren't tasks, they're choices I've been putting off. State the actual choice for each one.
4. KILL THESE - things I've been carrying that I'm honestly never going to do. Say so plainly.
5. What I mentioned in passing and then moved on from too fast. That's usually the important one.

Then ask me the 5 questions you'd need answered to make this list actually accurate.

How it works

Say it, don't type it. Nobody types their way to sixty loops. And block the time honestly, because a real catch-up pass is a morning, not a coffee break. The payoff is that you only owe yourself one of these a quarter if you run the smaller weekly version of it after.

What it does

A running note with one line per AI win: the date, what happened, whether it is worth repeating. Ten seconds to write. Then once a week you hand the log back to AI and ask what it sees. Your own week becomes the curriculum.

Why it's worth it

The agents who compound with this stuff are not smarter. They keep score. I log mine, and a weekly review of the log is what caught me rebuilding the same package by hand three weeks running; that chore is a routine now. The rule of thumb: anything that shows up three times gets promoted to a scheduled task.

The weekly mining prompt

Read my leverage log below. Three questions:

1. What did I do by hand three or more times that should become a scheduled routine?
2. Which win saved the most time, and what is the next-size-up version of it?
3. What am I doing that I should stop doing entirely?

My log (format: date, what happened, worth repeating?):
[PASTE YOUR LOG]

Keep the log inside your Chief of Staff project and it compounds twice: the weekly review gets smarter, and your chief of staff learns what actually works for you.

What it does

A simple sheet of your people: name, segment, birthday, one specific thing you know about them, last touch. Every Monday, AI tells you whose birthday or home anniversary lands in the next seven days and hands you the specific thing. Then you write the text.

Why it's worth it

My birthday was this month. The messages that meant something were the specific ones, the people who remembered a detail. That is the whole play. And here is the quiet part: the birthdays are the excuse. By the time the sheet is filled in, you have built the sphere database most agents pay a CRM for and never open.

The prompt

Help me build my sphere sheet. Make a table with these columns: Name, Segment (past client / sphere / lead), Birthday (stored as text, like 07-25), One specific thing I know about them, Last touch.

Now interview my memory. Go segment by segment and ask who comes to mind, past clients first. For each person, ask me for their birthday if I know it, and one SPECIFIC thing: their dog's name, their kid's team, the garden, the story of their deal. Do not accept "nice person" as an answer.

When we are done, set up a weekly check: every Monday morning, tell me whose birthday or home anniversary lands in the NEXT 7 DAYS, with their specific thing next to their name, so I have time to do something real about it instead of a day-of "HBD".

Two gotchas worth respecting. Store birthdays as text like 07-25 so nothing mangles them into dates. And the failure mode: automating this so well that you are no longer in it. The reminder is the machine's job. The message is yours, every time.

What it does

You hand AI one listing: the MLS report, the comps you pulled, the photos. It comes back with the whole kit, each piece in your branding with your license on it. A designed comps flyer for the seller. A neighborhood market report for your farm. A five-slide Instagram carousel with the caption already written. The same method handles the math-heavy pieces too, like this seller-credit rate flyer that shows what the same dollars do as a price cut versus a rate buydown.

Why it's worth it

This is the marketing every listing deserves and almost none get, because each piece used to be a designer's afternoon. Now every listing gets the full treatment the same day you take it. And a listing presentation that includes this kit is a different conversation with a seller.

See real examples

What it does

Every Tuesday, for each live escrow: what actually happened this week, what is due in the next seven days with dates, and a calm client email drafted in your voice. The open-escrow card handles day one. This handles days two through thirty-five, which is where clients actually get nervous.

Why it's worth it

Clients don't lose confidence over bad news. They lose it over silence. A weekly update that arrives before they thought to ask is the service, and it is also the referral story they tell later: "she kept us informed the whole way." This cadence is the core of the escrow tool I built; you can run the same pattern with nothing but a project per transaction.

The setup

Give each transaction its own project (see "stop re-explaining yourself" on the Week 2 page). Drop in the timeline, the key dates, and the important emails as they land. Then every Tuesday, run this in that project:

The prompt

It is Tuesday. For this transaction, using everything in this project:

1. What actually happened in the last 7 days? Docs in, contingencies removed, appraisal, loan milestones. Only things you can point to in this project.

2. What is due in the next 7 days? Give me a dates table: item, who owns it, due date. If a date comes from a contract document, tell me where you found it.

3. Draft the weekly update email to my client in my voice. Calm, plain English, no jargon. Lead with the one-line status, then what happened, then what happens next and what (if anything) we need from them. Two short paragraphs plus the dates list, no more.

If nothing happened this week, say so, and draft the "quiet week, here is what is coming" version instead. Never invent a status. If you are not sure whether something happened, ask me one question before drafting.

Check the dates table against your TC and your calendar for the first few weeks. AI reading a contract is a starting point, not a system of record. And it drafts only; you read it, you send it.

What it does

From the listing photos and the MLS Agent Full report, one conversation produces: a six-page branded listing report for the table (cover, the home, gallery, feature highlights, location, and a back page about you), a matching five-slide Instagram carousel, and three captions: public, agent-to-agent, and one casual version in your voice.

Why it's worth it

The flyer gets left on the counter. A bound six-page report gets carried out the door with your name on the back page, and the feed post matches the thing in their hands. I run this exact package for real listings; once the inputs are in, the whole set is one conversation.

The prompt

I am hosting an open house and want a print-ready package. I am uploading the MLS Agent Full report and the listing photos. My branding: [YOUR NAME], [BROKERAGE], [PHONE], [DRE #], brand colors [YOUR COLORS].

Build me:
1. A six-page listing report, print-ready: cover with hero photo and address, the home page (beds, baths, square footage, year built, lot, from the MLS sheet), a photo gallery page, feature highlight cards, a location and neighborhood page, and a back page about me with my photo and contact block.
2. A matching five-slide Instagram carousel in the same design language.
3. Three captions: one public, one written agent-to-agent, one casual in my own voice.

Rules: every fact comes from the MLS report I gave you; if you cannot point to it there, leave it out. Mirror what the listing highlights. Do NOT include any value estimate, price prediction, or "worth" language anywhere in any piece.

Verify every number against the MLS sheet before anything prints. And if it is another agent's listing you are sitting, mirror their highlights and show them the package before it goes anywhere. No valuations, ever; format the presentation, keep the judgment.

What it does

You hand it raw footage — a listing walkthrough, an open house, you talking through this week's market — and it hands back the whole package: a hook, a beat-by-beat cut plan, the words that go on screen for each cut, which frame to use as the cover, and the written caption with the call to action already in it.

Why it's worth it

My 8 year old sat next to me and described an app she wanted. A secret door, cherry blossom trees, a picnic. We built it by talking, and it was live on the internet by dinner. Then I asked for a Reel about it.

AI recorded the walkthrough, cut it to 20 seconds, wrote every word that appears on screen, made the cover image, and wrote the caption. I never opened an editor. Start to finish it was a conversation.

Your version: the listing video sitting on your phone right now that you haven't posted because editing it feels like a whole evening. It isn't anymore.

The prompt

I recorded a video of [a listing walkthrough / an open house / me explaining this week's market]. I'm going to describe it to you beat by beat, or upload it if you can watch it.

Give me back five things:

1. A hook for the first 2 seconds. Put the payoff in the first 5 words. No slow setup, nobody waits.
2. A cut plan: which moments to keep, in what order, and how long each stays on screen. Keep beats to 2-4 seconds.
3. The on-screen caption for each beat. 4-7 words each, absolute max. Nobody reads more than that in 2 seconds.
4. The cover: which frame to use, and the text that goes on it.
5. The written caption in my voice, ending with a line that tells people exactly who to send it to.

My audience is [who they are]. What I want them to do is [comment a keyword / DM me / save it].

The two things people get wrong

Change the caption at the same instant the picture changes. If the words fade in half a second late, the eye reads them against the wrong shot. Snap them on at the cut and people compare the words and the picture together.

Record way more than you need. Film 12 to 15 seconds of a moment even when the final cut is 4. Trim room is the whole difference between something that looks edited and something that looks like a phone video.

The version I run does the entire thing end to end — it drives the app, records the screen itself, cuts the beats, burns the captions and renders the cover with no phone involved. That's the deeper lane further down this page. You do not need it to get most of the way here.

What Claude Code actually is

The same Claude, moved onto your computer. The app can answer and draft; Claude Code can do: create and organize files, run work on a schedule while you are in the field, and build tools like the ones on this page. It is the advanced lane, and it is genuinely optional. We install it together in Week 5, and nobody is behind for waiting. This guide exists for the people who are caught up and want the head start.

Before anything: two gates

First, it needs a paid Claude plan (Pro or above); the free plan does not include it (as of July 2026). Second, know the easy exit: if the word "terminal" makes you want to close this tab, install the Claude Desktop app instead and use the Code tab inside it. Same power, no terminal. The steps below are the full version.

Windows, in PowerShell

irm https://claude.ai/install.ps1 | iex

Mac, in Terminal

curl -fsSL https://claude.ai/install.sh | bash

When anything snags

Open the Claude app and say: "I am installing Claude Code on Windows and hit this error," then paste the error. That is not cheating; that is the whole point of this class, and it is exactly how I fixed mine. First thing to try once you are in: ask it "what could you do for a realtor from this folder?"

What it does

Most people use AI like a stranger they meet every morning. New chat, same forty seconds of "I'm a realtor in the San Gabriel Valley, I work with first-time buyers, my brokerage is..." That's the tax you're paying, and you can stop paying it.

Claude reads context in three layers, and they stack. Before it answers you, it's already pulled from all three.

Why it's worth it

The difference between an agent who says "AI is fine, I guess" and an agent who runs their business on it is almost never the prompting. It's whether the AI knows anything about them. A generic assistant gives generic answers. Nine minutes of setup here changes the output on every prompt on this page.

And the second layer matters more than people expect. Once your projects and your memory are real, you stop writing long prompts. "Draft the follow-up" starts being enough, because it already knows who, about what, and in what voice.

The projects to build first

Don't build twelve. Build the three where you repeat yourself the most. For most agents that's Listings & sellers, Buyers & transactions, and Marketing & content. Add a Recruiting or Farm project later if that's your business.

The prompt

Here's the part almost nobody does. Don't write the project instructions yourself. Make Claude interview you and write them. Start a chat inside the new empty project and paste this.

I'm setting up this Claude Project for one area of my real estate business: [LISTINGS AND SELLERS].

I want you to write the custom instructions for this project. But don't guess, and don't write them yet.

First, interview me. Ask me one question at a time, and use my answers to decide what to ask next. Dig into:
- What I actually do in this part of my business, start to finish
- Who the people are: my clients, my vendors, who I hand things off to
- What I repeat over and over that you could take off my plate
- My standards and my non-negotiables, the things I'd correct you on
- How I want things formatted when you hand them back to me
- What's off limits, what I never want you to say or promise on my behalf

Keep asking until you genuinely have enough. Don't stop early to be polite, and don't ask me things you could infer from what I already said.

When you're done, write the project instructions and show them to me. Write them as instructions to you, not as notes about me. Then tell me what files I should upload into this project to make you better at it.

How to run it

One more thing: reference an old chat

You don't have to rebuild context you already built. In a new chat you can point Claude back at an old one and it'll pull the thread forward.

Look back at our earlier conversation about [the Arcadia listing / my open house follow-up system / the counter offer we reviewed] and pick up where we left off. Summarize what we decided so I know you've got the right one, then keep going.

Works best when the old chat lived in a project, because then Claude knows exactly which pile to look in. Another quiet argument for setting the projects up.

What it does

One project that holds your whole world: your buyers, your listings, who you owe what and by when. You talk to it in plain English after every showing and every call. It files what you tell it, surfaces what is due, and tells you what to work on next. Every other card on this page is an employee. This is the manager.

Why it's worth it

This is how I actually run my business. The morning brief, the inbox triage, the follow-up radar on this page: mine all report to one chief of staff that knows my priorities, my people, and my open loops. The change isn't any single trick. The burden of remembering what's open moved out of my head and into a system that never forgets a commitment.

The setup prompt

You are going to be my chief of staff. Before you start, interview me. Ask one question at a time and wait for my answer:

1. What does my business look like right now? Active buyers, listings, escrows, anything mid-flight.
2. Who are the people I owe something to right now, and what do I owe them?
3. What are my top three priorities for the next 90 days?
4. What slips through the cracks most often in my week?
5. What would the perfect Monday morning update from you look like?

Then write Project Instructions for this project that make you my chief of staff:
- Track my open loops: every commitment I mention gets filed with a due date. Never let one disappear.
- Track my people: clients, leads, sphere, each with the last touch and the next one.
- When I tell you something happened, file it and confirm in one line.
- When I open this project, do not wait for me to ask. Tell me what you would work on first and why.
- Never make me re-explain my business. Start every answer from what you already know.

Give me the instructions block to paste into the project settings. Then start the job: ask me for today's open loops.

How to set it up

What it does

Reads your unread email, tells you what each person actually wants in one line, sorts everything into buckets by urgency, and drafts replies for the ones that need one. Two prompts: a one-time setup that teaches it your world, then a daily one that uses what it learned.

Why it's worth it

Mine runs on its own, multiple times a day, and I never start it myself. It drafts a reply to every email that needs one, sitting in Gmail waiting on me. Because it's trained on my voice, the drafts are close enough that I skim and hit send.

The run that convinced me: 22 unread, 2 needed me that day, 4 that week, 16 didn't need me at all. The drafts are what changes your morning. You're editing instead of staring at a blank reply box.

Step 1 — teach it your world (once)

I want to set up an email triage system that runs every morning and drafts my replies.

Before we build it, interview me. Ask me these questions one at a time, and wait for my answer before moving to the next one:

1. Who are the people whose emails always matter, no matter what they say? (clients, your broker, escrow, lenders, your team)
2. What deals or addresses are live right now that I should be watching for?
3. What kinds of email do I get in a normal week? Walk me through the categories.
4. Which of those actually need me, and which are just noise I can ignore?
5. What kinds of email do I always want to answer myself, that AI should never draft for?
6. Is there anything that should get flagged as urgent even if it looks routine?

After I've answered everything, write me a "Triage Context" block that captures all of it — my VIPs, my live deals, my categories, my priority rules, and my hands-off list. Format it so I can paste it straight into a scheduled task.

Keep it tight. This is a reference card, not an essay.

Save what comes back. That's the part that makes this yours instead of a generic triage anyone could run.

Step 2 — the daily prompt (paste your Triage Context into the top)

[PASTE YOUR TRIAGE CONTEXT HERE]

---

Go through the last 24 hours of my inbox using the context above.

Sort everything into four buckets:
- NEEDS ME TODAY — someone is blocked or a deadline is real
- THIS WEEK — real, but not today
- WAITING ON THEM — I already replied, tracking so it doesn't go cold
- NO ACTION — newsletters, receipts, noise

For each one that matters: one line on who it's from and what they actually want. Not a summary of the email, the ask.

Then draft a reply to everything in the first two buckets, in my voice, saved to Gmail drafts as a reply on that thread. Include the ones that are just a quick yes — those are the ones that sit for three days.

If a draft needs a number or date only I would know, write it anyway and leave a [BRACKET] so I fill one blank instead of writing the whole thing.

Never send. Drafts only.

End with one line: how many needed me, how many didn't.

How to set it up

What it does

Assembles everything you'd otherwise open five apps to find, and hands it to you in one place. It reads your calendar and tells you what your day actually looks like, and it pulls the weather for where you'll be.

Why it's worth it

I used to open my weather app every morning. Now it's just there. That's the small version of the point. The bigger one is that I stopped rebuilding the same picture every day. Mine grew by conversation: I said "add the weather for my two cities" and two minutes later it was permanent. Start simple and let it grow.

The prompt

Every weekday at [TIME], give me a short brief on my day.

1. My calendar -- what's on it, who I'm meeting, and anything I need to prep for
2. Anything on my calendar where I should know something about the person before I walk in
3. The weather in [CITY] -- high, low, and what it's doing midday
4. Anything from yesterday I said I'd follow up on and haven't

Under 3 minutes to read. Short headers, bullets, no filler. If a day is light, say so in one line instead of padding it.

How to set it up

What it does

Goes and researches one subject you care about, every day, and hands you back what actually changed and where it got it.

Why it's worth it

Mine runs on SoCal housing and mortgage. Pick the thing you'd have to stay on top of anyway and hand it off. Your farm area. New construction going up near you. Rates. What the brokerage down the street is doing. The agent clients think of as the person who knows what's going on is the one who gets the referral.

The prompt

Every morning, research [TOPIC OR AREA] and give me a short brief.

Cover:
1. What actually changed in the last 24 hours, with the source
2. What it means for my clients, in plain English, not industry-speak
3. One thing in here I could use in a conversation, a text, or a post this week

Skip anything that's a rewrite of yesterday. If nothing real happened, say that in one line.

Once a week, add: one thing I should be watching that hasn't happened yet.

How to set it up

What it does

Reads tomorrow's calendar and hands you a page on each person — who they are, what they're doing publicly, what you've talked about, and what you promised them.

Why it's worth it

This is the one people react to hardest. You walk into coffee already knowing they just listed in Arcadia and that you last talked in March about the buyer who fell out. Fair warning: it will surface the follow-up you forgot.

The prompt

Look at my calendar for tomorrow. For every meeting with someone outside my company, give me a short brief:

- Who they are and what they do
- Anything public and recent worth knowing (their site, their listings, their socials)
- What we've emailed about before, and anything I said I'd follow up on
- Two things I could genuinely ask them about

Keep each person to under a page.

How to set it up

What it does

Looks across your conversations, finds who never replied to you and who you never got back to, ranks them by what matters, and drafts the follow-up.

Why it's worth it

Follow-up is the thing every agent knows they should do and almost nobody does. My last run surfaced 19 open loops I'd let go cold — and caught a typo in an email address that had been silently bouncing for weeks.

The prompt

Go through my sent email from the last 30 days. Find every conversation where I was waiting on a reply and never got one, and every one where they replied and I never answered.

Rank them by how much they matter to my business. For the top 10, write the follow-up message — short, warm, no guilt, easy for them to answer. Put them in my drafts.

How to set it up

What it does

You say everything that's in your head out loud, and it hands you back an order to work in. Not a transcript. A ranked list of what actually matters, the emails already drafted, and the things you said you'd do written onto your calendar.

Why it's worth it

Nobody ever has talker's block. Agents who say they hate writing don't hate writing, they hate starting. Talking is the one thing you already do all day.

This is the one I use most. Monday mornings, when I've got twelve things on my plate and no idea which one to touch first, I talk for five or ten minutes and let it sort me out. Not as a note taker. As a coach who tells me what to do first and what can wait.

The prompt — when the week is on top of you

Act as my business coach. I'm going to talk for five to ten minutes and dump everything on my plate right now — deals, open loops, people I owe a reply to, and the stuff I keep pushing to tomorrow.

I have [how many hours] to work today.

Don't just organize it back at me. Tell me:

1. The 3 things that actually move my business this week, and why the rest can wait
2. What I should drop, delay, or hand to someone else
3. Anything you can hear me avoiding

Then help me knock out number one. If it's an email or a text, draft it in my voice.

The prompt — right after a showing or a meeting

I'm going to talk for two minutes about what just happened. Don't clean it up into corporate language. Turn it into:

1. A ranked to-do list of what I actually need to do
2. A short update email to my client, in my voice
3. Anything I said I'd do that I should put on my calendar

How to run it

What it does

Your phone becomes a translator for a real conversation. You speak English, it speaks their language. They answer, it tells you what they said.

Why it's worth it

I used this in an actual face-to-face conversation with someone who didn't speak English. In our market that's not an edge case, it's Tuesday. The alternative is losing the conversation or waiting on the adult child to translate — and the parents are usually the ones with the money.

The prompt

I'm sitting with someone who speaks [language]. Be our translator.

I'll speak English, you say it in [language]. They'll answer, you tell me in English. Keep it natural and conversational, not formal. Don't add commentary, just translate.

How to run it

What it does

You photograph the whole stack of cards at once and talk through who each person was. It pulls every name, email and phone into a clean list, then writes a personal follow-up email and a text to each one, in your voice, referencing the specific thing you said about them.

Why it's worth it

Doing this by hand is the reason most cards die in a jacket pocket. Typing in twenty names, twenty emails, then writing twenty messages that don't sound copy-pasted is an hour of work, easily more. So it doesn't happen, and the whole reason you went to the event evaporates.

By voice it's five or ten minutes. You talk, it writes, everything lands in your drafts, and you review and send. Twenty people who met you once now hear from you the same night, each one referencing something real.

The prompt

Here are photos of the business cards I collected at [EVENT] tonight. I'm going to talk through who each person was.

First, read every card and pull the details into a clean list: name, company, title, email, phone. Flag anything you can't read clearly so I can fix it instead of you guessing.

Then, for each person, using my voice skill:

1. Write a short follow-up EMAIL that references the specific thing I said about them. Warm, brief, no pitch. Save it to my Gmail drafts.
2. Write a short TEXT for the same person. Shorter and more casual than the email. Just the message, ready to send.

Rules:
- Every message references something real about that person. If I didn't say anything specific about someone, write a warm general note instead of inventing a detail.
- No two messages should open the same way.
- Don't send anything. Drafts only, and give me the texts in a list I can work through.

How to run it

What it does

Sorts the report into what actually matters, what's worth asking about, and what's normal for a house that age — then writes the email that walks your buyer through it.

Why it's worth it

The inspection is where deals die from panic. Most of what scares a buyer is routine. Separating real problems from noise, in language they understand, is most of the job — and it's the moment they decide whether they trust you.

The prompt

Here's the inspection report for [address]. My buyer is a [first-time buyer / investor / family] and they're going to be overwhelmed by this.

Break it into:
1. Things that actually matter — safety, structure, water, systems at end of life
2. Things worth asking about but not alarming
3. Normal wear that every house this age has

Then write an email to them in my voice that walks through it calmly and tells them what I'd suggest we do next. No jargon.

How to run it

What it does

Takes your exported comp data and lays it out as something you'd be proud to hand across a kitchen table.

Why it's worth it

Every agent at that listing appointment hands over the same grey spreadsheet. The data isn't your differentiator — how seriously it looks like you took it is.

The prompt

Here's my comp data for [address] as a spreadsheet. Build me a clean, professional one-page summary I can hand a client — my branding, easy to read, the comps laid out so the comparison is obvious at a glance. Give me a version I can save as a PDF.

Don't add any conclusion about what the home is worth. That's my conversation to have.

How to run it

What it does

You upload the executed purchase agreement and every counter. It reconciles the final terms across all of them, then drafts an email to each party in the transaction, sending each one only what's relevant to them.

Why it's worth it

Opening escrow is the most chaotic hour in the deal. Buyers, listing agent, both TCs, escrow, title, lender, inspectors, home warranty, insurance, that's eleven people who each need something different, pulled from terms scattered across a contract and two counters. By hand it's an hour and it's where things get missed, usually the detail buried in the second counter. By voice it's a few minutes and everyone hears from you the same afternoon.

The moment that lands it: the final price often isn't in any single document. It only exists after you reconcile all three. When AI surfaces the right number, you can see it read everything, not just the last page.

The prompt

We just opened escrow on [ADDRESS]. Attached are the purchase agreement and both counters.

Read all three and reconcile the final terms, price, credits, dates, possession, everything that changed across the counters. Show me the reconciled terms in a short table first so I can confirm them.

Then, using my voice skill, draft an email to every party: buyers, listing agent, both transaction coordinators, escrow, title, lender, general inspector, termite inspector, home warranty, insurance.

Send each party only what's relevant to them. The lender needs loan amount and credit structure; title needs the parties and property; inspectors need access and the deadline; nobody outside the deal needs the buyer's financials.

Also give me a one-page escrow summary for my own file, and a short text I can send the buyers congratulating them.

Save the emails to my Gmail drafts. Don't send anything.

Practice on the demo deal

This is the exact transaction I ran live in Week 2. Fake deal, fake people, real paperwork structure. Download all three, upload them into one chat, and run the prompt above. You'll see the whole thing happen on a deal where nothing is at stake.

Or grab the whole folder here. Every page is stamped as a training document. Don't use these as contract templates, they're built to teach, not to transact.

Check its work

Here's the point of the demo. The final purchase price is $1,285,000, and that number appears in only one of the three documents. Not the contract. Not the first counter. If you'd read the purchase agreement you'd be off by $17,000, and if you'd read the seller's counter you'd be off by $10,000 in the other direction.

Two more to check it on: the seller credit is $15,000, ten of it a permanent rate buydown, five toward closing costs. And possession is a week after close, not at close, because of the rent-back that got negotiated in the last counter. That rent-back detail is exactly the kind of thing that gets missed by hand and then blows up two days before move-in.

If your AI comes back with a table that gets those three right, it read all three documents. If it doesn't, tell it so and make it reconcile again. That habit of checking the summary before you trust the drafts is the actual skill here.

How to run it

What it does

All your contacts live in it, organized into lists. When you want to reach out, you just tell it who and what: "Text my open house list and the Chens, invite them to Saturday's broker open at 143 Maple." It figures out the audience, shows you exactly who it picked so you can add or drop anyone, writes a different message for each person, and you send them from your own phone one tap at a time. It is not a blast. Every message passes in front of you before it goes.

Why it's worth it

Nobody skips follow-up because they don't care. They skip it because fifty personal texts is a two hour job, and a little bit of an emotional one. This turns it into about five minutes, and you can talk instead of type. Say who and what out loud while the coffee brews, review the drafts in the Starbucks line, tap them out between sets at the gym. And because every text goes from your number, one at a time, it reads like you thought about the person. That's the only kind that gets answered. Load your contacts once, a spreadsheet, pasted names, or your phone's exported contacts, and it becomes the one place your sphere actually lives.

Where to go

What it does

You upload the inspection PDF. It comes back with what actually matters, what's worth asking about, and what's normal for a house that age. Then it writes the email to your buyer, in their language.

Why it's worth it

This is the inspection prompt above, except built. It knows what an FHA or VA appraiser is going to flag before it kills your escrow, and it knows what a California insurer refuses to write. Those are the two things that quietly blow up deals and neither one jumps out at you in a seventy page report.

Where to go

What it does

Takes a presentation from nothing to finished: outline, slide deck, and then the part almost nobody knows about — it listens to your rehearsal and tells you what to fix.

Why it's worth it

I built this bootcamp with it. I took Tony Robbins' AI bootcamp transcripts and slides, dropped them in a folder, voice-dumped my own context — who my audience is, what I actually wanted them to walk out with — and had it build a realtor version. Weeks of deck work became days.

But the rehearsal loop is the part that changed things. Practice used to have no feedback. You run through it, you feel okay about it, and you have no idea what actually landed. Now every run-through gets scored.

Your version of this: a homebuyer seminar. Build it once, record it, and it becomes an evergreen lead magnet you send to every buyer for the next two years. Same for a listing presentation or an investor talk.

Step 1 — the outline

I'm building a [90-minute homebuyer seminar / listing presentation / investor talk].

My audience is [who they are and what they already know]. What I want them to walk out with is [the one thing]. What I want to happen afterward is [book a consultation / sign a listing agreement / join my list].

I'm going to talk for a few minutes about everything I know on this and everything I want to cover. Don't clean it up yet — just listen.

Then build me an outline: what to open with, the order of the sections, how long each should run, and where the interactive moments go. Tell me honestly what I should cut.

Have material from someone else — a training you paid for, a webinar you liked? Upload the slides or transcript and say "use this as the structure, but rebuild it for my audience." That's exactly what I did.

Step 2 — the slides

Turn that outline into a slide-by-slide structure I can hand to a slide generator.

For each slide: the headline, two or three short supporting lines, and what visual belongs there. Keep every slide sparse — I'm doing the talking, the slide is just the anchor. Tell me the exact number of slides.

Give it to me as one block I can copy and paste.

Paste that into NotebookLM — it makes the best-looking decks of anything I've tried. Download as PowerPoint, then bring it back to Claude for feedback and iterate. Two or three rounds gets it sharp.

Step 3 — the rehearsal coach (this is the one)

Here's the transcript of me practicing this presentation out loud.

Be honest, not encouraging. Tell me:
- What ran long and what I should cut
- Anything I said that was inaccurate, overstated, or that I can't back up
- Where I lost energy or rambled
- Where the audience would tune out
- What actually landed and I should keep

Then give me the specific changes to make before I run it again.

How to run the whole loop

Then make it evergreen

Once it's good, have AI build a resource page for it — the handouts, the links, the recording. Host it free on Vercel or your own site. Now anyone who missed it can catch up, and anyone you meet next year gets the link.

That's the difference between a seminar you gave once and an asset that keeps generating conversations. This page you're reading is that.

What it does

You screenshot your profile, say what you're going for, and go back and forth until it's right. No counting characters, no guessing where the line breaks.

Why it's worth it

I stopped retyping mine into the app just to see how it wrapped. We went back and forth about six times in ten minutes and landed somewhere I wouldn't have reached alone. Screenshot, don't retype — that's the whole shortcut.

The prompt

Here's a screenshot of my Instagram profile. I'm a [what you do] in [market] and I want people to immediately understand [what you want them to get].

Give me 3 different versions of the bio. Respect the character limit and the fact that only about 4 lines show before it truncates. Make the first line the one that stops someone.

How to run it

What it does

Researches current posting-time data from real sources and splits the answer by content type, instead of repeating advice everyone's been copying since 2019.

Why it's worth it

I asked because my content partner and I disagreed. It came back with Reels 6–9pm, teaching content 7–9am — and then told me posting time was the fourth most important lever, and my real problem was the first five words of my hook. That was worth more than the answer I asked for.

The prompt

Research the current best posting times for real estate content on Instagram, using real sources, not general advice. Split it by content type — Reels vs carousels vs teaching posts. My audience is [who] in [market].

Then tell me honestly: how much does timing actually matter compared to everything else I could fix?

How to run it

What it does

Takes one piece of real raw material and turns it into everything you'd otherwise write separately.

Why it's worth it

The reason people stop posting isn't laziness, it's the blank page five times a week. Start from one real thing instead. One open house, one deal that almost fell apart, one thing a client said that stuck with you.

The prompt

Here's a transcript from [an open house walkthrough / a voice note about something that happened this week].

Turn it into:
1. An Instagram caption in my voice
2. A carousel outline, 5 slides, one idea per slide
3. A short email to my sphere
4. Three post ideas that come out of the same story

Same voice throughout. Don't make it sound like marketing.

How to run it

What it does

This one's a workflow note, not a prompt. I write the strategy, the hook and the caption in Claude, then generate the actual carousel images in ChatGPT or Gemini, because right now they produce more cinematic visuals.

Why it's worth it

Anyone telling you one tool is best at everything is selling something. Using two on purpose isn't a failure, it's knowing what each is good at. And it changes — check again every few months.

The split

What it does

You describe your business and what you want people to feel, and AI designs and builds the site. Not a template with your headshot dropped in. joshuasun.co is the proof — the main site, plus /ai for the bootcamp and /rebuild for the Altadena project. They scroll like a film instead of a page.

Why it's worth it

Every agent has the same site as every other agent. This is the fastest way I know to not look like everyone else. Full disclosure: I'm not a developer. I got there by directing tools, and no AI one-shots a finished site — expect to regenerate and adjust. Budget time for that, not just money.

The prompt — the part that actually matters

I want to build my own website and I'm not a developer. Before you build anything, I'm going to give you everything about my business.

[Then talk for as long as it takes. Who you serve. What you actually do for them. What you want someone to feel in the first five seconds. What you want them to do next. Don't summarize — more context is better.]

Then:
1. Ask me whatever you still need to know
2. Show me the design direction before you build — layout, colors, fonts, the feel
3. Build it once we agree on that

Don't hand me a generic agent template.

This is the whole difference. I gave it one source of truth and fed it my thinking, not a one-line prompt. The site came out close to what I actually meant because I front-loaded the why, not the what.

The tools

The full walkthrough

I wrote the whole thing up, tool by tool, including what it cost and where I got stuck: How I Built a Website With AI — The Playbook.

See a live example → A sample built for a fictional agent, so you can click through before you build your own. Sample data only — nothing here is a real business.

What it does

One page: your escrows with the next deadline on each, your listings and where each is in launch, this week's new leads, the follow-ups due, the conversations you have started. Built as a single HTML file you save and open in your browser. No hosting, no login, no monthly software.

Why it's worth it

Your CRM shows you your CRM. Nothing shows you your business. Mine started as one conversation and now it is the first tab I see every morning: what is running, what is stalled, what needs me. The design rule that makes it honest: the numbers you control get the big font (calls, doors, posts, open houses), the results render small (closings, volume). You stare at the inputs; the outputs follow.

The prompt

Build me a one-page business dashboard as a SINGLE self-contained HTML file I can save and open in my browser. No hosting, no accounts, everything in one file.

Interview me first, one question at a time:
1. If I could only see five things about my business every morning, what should they be? Help me think: active escrows with their next deadline, listings and their launch stage, new leads this week, follow-ups due, conversations started this week.
2. Which of my numbers are lead measures, the ones I control weekly (calls, doors knocked, posts, open houses hosted)? Those get the biggest type on the page.
3. Which are lag measures (closings, volume)? Those render smaller, on purpose.
4. My brand colors, and whether I want it bright or calm.
5. Anything money-related that should hide behind a "demo mode" button before I ever screen-share this.

Then build the file: clean, bold, readable from across the room. Include the demo-mode button. One honesty rule: never show a number I did not give you. If a section has no data yet, show "not wired yet" instead of inventing something.

I will tell you what to change and we will go back and forth until I actually enjoy looking at it.

How to set it up

The moment real dollars are on this page it is orange-zone: build the demo-mode button before you ever present or screen-share, not after. A screenshot travels further than you plan for.

What it does

A nudge to hit record before a meeting. A nudge to voice-dump right after, while it's fresh. The transcript filed somewhere searchable. And a watchdog that surfaces the person again when they go cold.

Why I built it

I kept having great conversations and then losing them. Not forgetting the person — forgetting what we actually said, which is the part that makes the next conversation good. Four small reminders chained together beat a CRM I'd never open.

Where to start

You don't need Claude Code for the first version. Start with the meeting brief above, and add a phone reminder after each meeting that just says "voice dump now." That's 80% of the value with none of the setup.

What it does

I put raw clips in a folder, describe what I want out loud, and get back edited, captioned, branded content. Revisions are spoken, not clicked.

Why it matters

It removed about 90% of the editing work. Not 90% of the thinking — I still decide what the piece is about and whether it's any good. But the part where you drag clips around at 11pm is gone. This is the single biggest reason I'd tell you to look at Claude Code.

Where to start

We'll go through it properly in Week 5. Until then, the version you can do today: record the footage, and use "one real moment becomes a week of content" above to handle everything except the edit.

What it does

Keeps a registry of everything running on a schedule and checks whether each one is actually still doing its job.

Why it matters

Once you have five things running automatically, the failure to fear isn't a crash — it's silent degradation. Mine caught five automations that had all fired nine hours late, in one burst, with no error message anywhere. I'd never have noticed.

Where to start

Nowhere yet. This only matters once you have things running. It's here so that when you do, you know to check on them.

What it does

Reads about fifty of your own real emails and writes down how you sound. How you open, how you close, your rhythm, your words, and the things your voice never does. You save that once, and from then on everything AI writes for you comes out sounding like you instead of like a press release.

Why it's worth it

This is the one that makes everything else on this page usable. Drafted replies, follow-ups, captions, none of it saves you a minute if you have to rewrite it. You already wrote the training material. It's sitting in your Sent folder right now.

The prompt

I want you to analyze my writing voice and create a reusable Voice Skill file I can use to make AI write like me.

## STEP 1 - GO FIND MY WRITING

Don't ask me to paste examples. Search my Gmail Sent folder and pull my actual writing.

What to look for:
- Roughly 50 of my LONGEST, most substantive sent emails. Real messages where I explained something, persuaded someone, handled a problem, or wrote at length.
- NOT one-line replies. Skip "sounds good," "thanks, see you then," "got it," and anything under about three sentences.
- Variety matters: client emails, problem-solving emails, intro and follow-up emails, anything longer and more personal.

CRITICAL - skip anything AI wrote:
- If an email reads like AI wrote it, exclude it. That's not my voice, that's ChatGPT's.
- Tells: em dashes as connectors, "I hope this email finds you well," "I wanted to reach out," "Let's dive in," perfectly balanced three-item lists, no contractions, no typos, uniform paragraph lengths.
- If I started using AI at some point, reach back to BEFORE that. Go back one, two, even three years. Older writing beats recent AI-assisted writing.
- Tell me what date range you used and why.

Before you analyze, report back: how many emails you found, the date range, and how many you excluded. If you found fewer than 15 substantive ones, STOP and tell me so I can paste some manually.

## STEP 2 - ANALYZE

For every email examine:

Structure: How do I open? How do I close? How long are my paragraphs? Headers, bullets, or flowing prose?

Rhythm: Short sentences, long, or mixed? Do I use fragments? Do I vary length for effect?

Word choices: Contractions? Slang? Shortcuts like "asap" or "thru"? Industry language? What's my register - friend, client, colleague?

Personality: Humor, empathy, authority, warmth, directness? How do I handle bad news, disagreement, or an upset client?

What I DON'T do: What punctuation do I avoid? What phrases would sound fake in my voice? What tonal registers never show up?

EVIDENCE RULE: Every pattern must be backed by a direct quote from my actual emails. In 2+ emails it's a rule. Only once, it's a tendency - say which. Never invent or paraphrase. Quote me.

## STEP 3 - BUILD THE FILE

Create this exact structure. It must work standing alone as a Skill file.

CRITICAL: the file MUST begin with the YAML frontmatter block below, exactly as shown, starting on line 1. Three dashes, the two fields, three dashes. Nothing before it, no code fences around it, no blank line above it. If this block is missing or malformed, the file will be rejected when I try to upload it as a Skill.

---
name: [my-first-name]-voice
description: Write in [My Name]'s voice. Use whenever writing an email, text message, Slack message, social post, or any other communication on my behalf.
---

# [My Name] - Voice Skill

You are writing as [My Name]. [One sentence about what I do and who I write for.]

Extracted from my actual sent email across [date range]. Internalize these as a whole portrait, not a checklist. The goal is writing I'd read and think, "yeah, that sounds like me."

## Core Identity
[One paragraph. Not personality adjectives. Describe the RELATIONSHIP between me and my reader. What should they FEEL?]

## Voice Principles
[5-8 principles. Each one:]

### [Number]. [Principle Name]
[2-3 sentences on what it means and why it matters.]
The rule: [One actionable sentence an AI can follow.]

| Instead of | Write |
|---|---|
| [Generic AI-default] | [How I actually write it - real quote] |
| [Another generic] | [My version] |

## Signature Patterns

### How I Open
[3+ actual opening lines. Then the pattern.]

### How I Close
[3+ actual closing lines, then the pattern, including how I sign off.]

### Transitions
[The actual transition words I use, pulled from my emails.]

### How I Handle Hard Things
[Bad news, disagreement, chasing someone gone quiet, an upset client. 2-3 real examples and the pattern. This is where voice matters most.]

## Anti-Patterns: What My Voice NEVER Does
[5-10 things that would make writing NOT sound like me. Each one:]

### No [Pattern Name]
[One sentence: why this breaks my voice.]

| Instead of | Write |
|---|---|
| [Wrong version] | [Correct version for my voice] |

Look for: punctuation I avoid (em dashes? exclamation points? semicolons?), filler that'd sound fake ("It's worth noting," "Let's dive in," "I hope this email finds you well"), structural patterns I never use, tonal registers that don't fit, and AI tells ("I'd be happy to help," "Great question!").

## Formatting & Language Rules
- Paragraph length: [typical sentence count]
- Sentence style: [short? long? mixed?]
- Contractions: [always/sometimes/never, which ones]
- Shortcuts: [do I write "thru," "asap," "ppl"?]
- Headers and bullets: [when I use them vs write it out]
- Bold/italic: [what gets emphasis and why]
- Emoji: [frequency, placement, which ones if any]
- Sign-off: [exactly how I end an email]

## Format Adaptation

| Format | What stays the same | What flexes |
|---|---|---|
| Text message | [core principles] | [length, formality] |
| Quick email reply | [core principles] | [what changes] |
| Long client email | [core principles] | [what changes] |
| Something public | [core principles] | [what changes] |

## The Voice Check
[3 specific questions to verify a piece sounds like me. Based on patterns you actually found, not generic.]

## RULES FOR YOUR ANALYSIS
- The "Instead of / Write" tables are the most important part. At least 15 across all sections.
- Pull ACTUAL phrases as evidence. Never paraphrase or invent.
- Anti-patterns matter as much as principles. AI defaults to generic unless told not to.
- Write it as instructions TO an AI: "Write as..." "Never..." "Always..." - not "The author tends to..."
- Output a reusable file that works immediately if saved as a skill.
- Start the file with the YAML frontmatter block. This is not optional. A file without it cannot be uploaded as a Skill.
- Give me the finished file as a downloadable .md file, not just text in the chat.

Now the part that actually matters

Whenever you're writing anything for me — an email, a text, a Slack message, a social post, any communication — use my voice skill.

That line is the whole reason to make it a Skill instead of dumping the profile into your Instructions. Your voice profile loads only when it's actually relevant to writing something, instead of being read on every single chat. The rest of the time it stays out of the way.