When a new listing comes in from your CRM or MLS feed, an n8n workflow has Claude describe the photos, draft the MLS description, social captions, a buyer-list email, and a tour narration script, park anything low-confidence for a person, and wait for your Approve in Slack before posting
An n8n template by Max (Agentmelt blueprints), published November 12, 2026, receives a new listing by webhook from a CRM, a Google Form, or an MLS feed change, with address, price, beds and baths, square footage, features, agent notes, and photo URLs. Three Claude steps run in order: one describes each photo (for example a kitchen with a quartz island or a west-facing deck) so the copy references what buyers will see; one drafts the MLS description within the MLS character limit in the agent's voice, leading with the strongest feature, with a fair-housing check for protected-class language; and one drafts Instagram and Facebook captions, a just-listed story script, a buyer-list email with the three best photos, and a 60-second virtual-tour narration script. After every Claude step a 'Needs a person?' gate parks low-confidence output instead of passing it on. The rules in each step say facts are checked against the listing data with no invented square footage or school ratings, brokerage disclaimers and MLS-required phrases are inserted rather than left to the model, and the agent approves every asset. A compliance and brand check step is where the fair-housing wordlist, brokerage disclaimer, required MLS phrases, and price and feature consistency checks go. A Slack message then waits for Approve or Reject, approved posts go to a Facebook Page through the Facebook Graph API, and each listing is logged to Google Sheets. The template tells users to fill in its Config node first and keep test mode on for the first week.
At a glance
- Tools used
- n8nClaudeSlackFacebook Graph APIGoogle Sheets
- Category
- Listings
- Audience
- agent, team lead, listing coordinator
- Difficulty
- Builder
- Source
- Write real estate listing descriptions with Claude and Slack
Max (n8n creator, Agentmelt) · n8n template · Nov 12, 2026 - First step
- Import the template, fill in the Config (edit me) node with your name, brokerage, time zone, quiet hours, and Slack approval channel, add your fair-housing wordlist and brokerage disclaimer to the compliance step, keep test mode on, and send one past listing to the webhook.
Compliance note: The template's compliance step is a placeholder until you add your own fair-housing wordlist, brokerage disclaimer, and MLS-required phrases, so add them before going live, and keep the Slack approval on so nothing publishes without your review.
Suggested prompt
Written by this directory, inspired by the source. Replace the [bracketed] parts.
You draft the MLS description for this listing. Stay within [MLS CHARACTER LIMIT] characters, write in my voice, and lead with the strongest feature. Use only the listing data and photo descriptions below: no invented square footage, school ratings, or features. Do not describe who the home is ideal for, the neighbors, or any protected class. If anything needed is missing, say so instead of guessing, and return needs_human: true if you are not confident. Return JSON with summary, draft, confidence, needs_human, and reasons. Listing data: [paste] Photo descriptions: [paste]
Get 50 AI Prompts for Real Estate Agents, free
Subscribe to the Daily 5 and download the 48-page PDF right away. The prompts help with:
- Getting more leads from the contacts already in your CRM
- Faster follow-up, so no text or inquiry goes unanswered
- Less time on listing descriptions, paperwork, and showing instructions
- Copy-paste prompts tested by working agents, brokers, and landlords
- Five new real-world AI use cases every weekday morning in the Daily 5
Confirm by email. Unsubscribe in one click. Review every AI draft for accuracy and fair housing compliance before you send it.