Bundle · 9 skills

Real Estate Agents

9 Markdown skill files for listing agents and buyer agents. Fair-housing filter built into every output. The skills you reach for between showings — without leaking client information or making promises you can't keep. State-neutral with guided one-time setup, ~30 minutes.

9 skills~24 pages of Markdown

JurisdictionAny state — one-time setup

FormatPlain .md files (ZIP)

UpdatesLifetime, free

For agents who already use AI but worry about fair-housing slips, accidental dual-agency language, or invented MLS facts in listing copy. Every output is scrubbed against your state's commission rules and NAR COE before it reaches you. Skills ship state-neutral with a guided setup to plug in your state's specifics.

What's inside

9 skills for the working agent.

listing-description

MLS-safe listing copy

Draft listing descriptions that won't slip into fair-housing trouble — the language filter flags steering, exclusion, and protected-class signals.

cma-narrative

CMA narrative writer

Turn comp data into a clear narrative for a seller meeting. Cites actual comps; flags when the data is too thin to support a conclusion.

offer-analysis

Multi-offer comparison

Side-by-side breakdown of competing offers with net-to-seller math and contingency-risk flags. Does NOT recommend — the call stays yours.

buyer-onboarding

Buyer intake & expectations

A first-meeting script that gathers price, area, and lifestyle needs without straying into protected-class territory.

followup-sequences

Buyer & seller followup

30/60/90-day followup templates for buyers post-close and for sellers post-listing-expiry. Personalized, not spammy.

farming-letters

Neighborhood farming copy

Quarterly farming letter drafts grounded in actual neighborhood data — no invented "trends" or fabricated price changes.

open-house-recap

Open-house followup pack

Same-day recap to the seller plus follow-up emails to attendees. Tone-matched, no pressure tactics.

objection-response

Buyer/seller objection drafter

Drafts responses to common objections (price, timing, repairs) that acknowledge the concern rather than steamrolling it.

social-post-pack

Weekly social pack

5 posts a week pulled from your actual activity — no generic "real estate tips" filler, no AI tells.

Sample skill

Here's what one looks like.

Every skill is a plain text file. The settings at the top tell the AI exactly which rules to follow on every reply. The rest of the file is the actual job instructions.

---

name: listing-description

jurisdiction: [your state]

version: 2026.05

guardrails:

  human_in_loop: true

  fair_housing_filter: strict

  do_not_use_phrases: [walking distance, family-friendly, safe neighborhood, exclusive]

  cite_or_flag: [MLS-verifiable facts only]

  do_not_invent: [square footage, year built, school ratings, HOA amenities]

  escalation_triggers: [protected-class signals, exclusionary language]

---

# MLS-Safe Listing Description

You are drafting an MLS listing description.

Pull only facts from the property data the user provides.

Run every draft through the fair-housing filter before returning...

Who this is for

Three workflows agents run every week.

Writing a listing without fair-housing slips

New listing goes live tomorrow. Drop the property facts into Claude with listing-description and get copy that's already filtered for steering language.

Comparing multiple offers under pressure

Three offers in, seller wants a summary by noon. Paste the offers in with offer-analysis — get side-by-side math without the AI picking a winner for you.

Running a real follow-up cadence

Buyers from your last 90 days. Use followup-sequences for individual notes that reference real details — not the generic "thinking of you" everyone else sends.

Common questions

Before you buy.

Will this make my listings sound like AI wrote them?

No. The skill is built specifically to suppress AI tells — "boasts," "nestled in," "stunning," em-dash floods — and to use neutral, verifiable facts. Output reads like an experienced agent wrote it.

Will the fair-housing filter actually catch real slips?

Yes. The filter flags steering language, protected-class signals, and the everyday phrases that put agents in trouble ("great for families," "perfect for retirees," "walking distance to schools," "safe neighborhood"). Every output shows you what got caught and why, so you learn the patterns over time.

Can I customize them?

Yes — they're plain Markdown. Open in any editor, modify, save. Swap in your own example listings, adjust the voice to match your style, layer on guardrails specific to your team.

How are updates delivered?

Email. When NAR COE is amended, federal fair-housing guidance changes, or the workflow skills are improved, you get a link to the revised ZIP. No portal to log into.