Skip to content

Free class · Get found

How to build real estate landing pages with AI.

Most agents open a chat, type "write my bio," and get copy that could be any agent in any city. This class flips the order. You build one Agent Context folder first, then ship three real pages from it: one for you, one for a listing, and one for a single client.

Ryan Wanner, founder of AI AccelerationInnovation Lab North, Compass Lakeside, HendersonvilleEvery class on this page is free to read and watch. No account needed.
Full class recording · 1 hr 38 min · recorded liveWatch on YouTube ↗

What you'll learn.

  • 01Why your edge is your own context and judgment, not whichever model is newest this month.
  • 02How to set up an Agent Context folder: my business, my clients, and my deals, as plain text files.
  • 03How to write guardrails the AI follows every time: Fair Housing, privacy, source rules, and banned phrases.
  • 04How to build a page about you, a page for one listing, and a private page for one client from the same folder.
  • 05How consent rules inside a client file change what the AI is allowed to say about that person.
  • 06Why every unverified fact gets a VERIFY mark and stays off the live page until you confirm it.

Part 1

The model is not the edge. You are.

Models change fast. The tool you rely on this month can be restricted, repriced, or replaced next month. If your whole approach depends on holding the newest model, you lose it the day the model moves. Your understanding and your context go with you to any tool.

Strong model, no context

Opens the most capable model available and types "write me a listing description." Gets generic copy that could be any agent in any market. Argues with it, then rewrites it by hand. When that model changes, they're back to zero.

Any model, real context

Loads their Agent Context first: who they are, their voice, their market, their guardrails. Gives even a cheaper model a real brief and gets copy that sounds like them and uses real facts. When a stronger model shows up, the same context makes it better too.

Get to the raw data.

AI hands you confident summaries all day. Break a claim down to something you can actually check, then build back up. If you can't get to the source and understand it, you're working from someone else's opinion. That's why every number on your page traces back to something real.

You own the output.

The AI won't be the one answering for a Fair Housing problem or a wrong square footage. You will. So you check the things that matter before anything goes live, and you don't over-check the low-stakes drafts that just need to get done.

Part 2

Your Agent Context is just a folder.

No software and no subscription. Three folders of plain text files inside one. Drop it into Claude, ChatGPT, or Gemini, tell it to read everything first, and it knows your business before you ask a single question. Every page, post, and email you build later comes out of the same center, so none of them start from a blank chat.

  1. 01

    agent context/

    The one folder you hand the AI before any task. Tell it to read everything in here first.

  2. 02

    my business/

    Who you are, how you sound, how you work, and your guardrails. Write it once. Every page starts here.

  3. 03

    my clients/

    One file per real person: their goals, how they decide, what never to say, and exactly what they've given you permission to use.

  4. 04

    my deals/

    One file per property: the facts, the angles, and a clear VERIFY mark on anything you haven't confirmed yet.

Consent lives in the client file.

Here's why the client folder matters. Picture a fictional agent with three client files, each with different rules. Every consent line defaults to no until the client says yes.

Client A: happy to be named

A past buyer who gave written permission to use their name and tell their story. Their page can be warm and specific.

Client B: past stays private

A buyer starting over after a hard few years. Their file lists what must never be mentioned, and the AI has to respect it.

Client C: no personal details

A seller who routes everything through an assistant and allows nothing personal in public. Their page stays strictly about the process.

Same prompt, three very different pages, because the file tells the AI what it may use. That's respect for your clients built into the system instead of left to memory. The deal folder works the same way for facts: anything unconfirmed is marked VERIFY, and nothing marked VERIFY gets published.

Part 3

Ship three pages in one sitting.

The move is the same every time. Tell the AI to read your context folder first, guardrails included, then give it one prompt. It writes the page in your voice from your real facts. Start with the page about you.

Page 1: you

Read every file in my "agent context/my business" folder first, then follow my guardrails for everything below.

Build a one-page personal landing page for me as a real estate agent. Output the full page copy plus a simple single-section HTML layout I can paste into a site builder.

Include, in my voice:
- A headline that says who I help and where, using my real market and specialty. No superlatives like "award-winning".
- A short "who I am" paragraph pulled from my about-me file.
- 3 specific reasons to work with me, stated as facts, not adjectives.
- One clear call to action.

Rules: use only facts in my files. Mark anything you're unsure of as VERIFY. Obey every Fair Housing and banned-phrase rule in my guardrails.
Page 1: you

Built from my business. The page buyers, sellers, and AI find when they look you up. Fix every VERIFY before it goes live.

Page 2: a deal

My business plus one deal file. Property type and neighborhood in the headline, 4 to 6 specific facts, objective sourced neighborhood notes, and one next step.

Page 3: a client

My business plus one client file. A private page for one person that answers what they said matters to them, within their consent rules.

Before you call it done.

  • Get all three pages live somewhere with a real link. Done beats perfect on this pass.
  • Clear every VERIFY mark. Publish facts, never guesses.
  • Send the page about you to 3 people who'll give you an honest read, and the client page to the one person it was built for.
  • Add another client file and another deal file this week, so the next page is even faster.

Free with a student account

Get the workbook and prompts, free.

Create a free student portal account to get the workbook, prompts, templates, and downloads for every class.

Create my free account →

Already a student? The same link signs you in.

  • Agent Context templates: my business, my clients, my deals
  • The guardrails template: Fair Housing, privacy, and source rules
  • All three page-build prompts: you, a deal, a client
  • Setup steps for Claude and ChatGPT
  • Pre-publish checklist and seven-day plan
  • Skill card: Build your MVP from Agent Context

Questions agents ask.

Do I need coding skills or a new website to build these pages?

No. Each prompt returns the page copy plus a simple single-section HTML layout you can paste into the site builder you already use. The first pass is about getting a real link live, not about design. Branding and polish come after the page exists.

Which AI should I use: ChatGPT, Claude, or Gemini?

Any of them. Your Agent Context is plain text, so it travels. In Claude you can load it into a Project. In ChatGPT you can paste your business files at the top of a new chat. The edge is your context and your judgment, not the brand of model.

What goes in the guardrails file?

Fair Housing rules, like describing the property and not the people who should live there. Privacy rules, like no client names, stories, or photos without explicit consent. Source rules, like never inventing square footage, schools, taxes, or stats. And the phrases you never want to see, such as "dream home" or "must-see".

Is it okay to put client details into an AI?

Only what you have a reason and permission to use. Each client file carries consent rules that default to no: name, story, photo, and a do-not-mention list. Keep private details off public pages entirely, and check your brokerage's policy before you load client information into any tool.

Why ship a rough page instead of a polished one?

Because a live page teaches you what a draft can't. MVP means minimum viable product: the smallest real version that works. Once it exists, you can improve it from your context instead of starting over. That's the next class in the series.

Research sources

Check the current rules, capabilities, and source basis.

  1. 01HUD: Fair Housing guidance for digital advertising ↗

    Government guidance on Fair Housing risk in digital housing advertising and delivery.

  2. 02NAR: 2026 Code of Ethics ↗

    Current professional standards requiring a true picture in advertising and prohibiting misleading images and manipulations.

  3. 03Google: Helpful, reliable, people-first content ↗

    Google's current questions for original value, first-hand expertise, authorship, sourcing, and satisfying the visitor's goal.

  4. 04Anthropic: Effective context engineering for AI agents ↗

    Why selected, high-signal context and durable notes outside the chat shape what a model produces.

  5. 05OpenAI: Why language models hallucinate ↗

    Why fluent output is not evidence, and why unverified facts need checking before you rely on them.

  6. 06FTC: Consumer privacy guidance for businesses ↗

    Collect and keep only the personal information you need, protect it, and describe your data practices accurately.

Product access, limits, local listing rules, platform policies, and laws can change. Re-check the linked authority before important or public work.

Keep building

Put AI to work in your real estate business.

AI Acceleration: Real Estate is a private, self-directed curriculum for practical AI work across content, communication, lead follow-up, CRM, client research, operations, and relationship management. Build it. Review it. Keep what works.

Explore the Skool community ↗

Follow Ryan + AI Acceleration

No ranking, revenue, closing, or time-saving promise. The classes teach the work and the review systems around it.

Record: innovation-lab-public-classes. Approved record hash: add45e5ab0ee95b80ad70a4ecb50c8b8317c6997119c13fd8a727225815aff73. Claims: public-classes-free-to-watch, public-classes-portal-unlock, public-classes-taught-live.