Find the deals already sitting in your lead list: give an AI one cash buyer's buy box and have it search your old seller leads for matching properties, drop the ones that don't pencil, and list who to call back first: AI agent prompt
Investors and wholesalers pay to generate seller leads, then most of them go quiet in the CRM. On the Collecting Keys podcast (write-up dated Oct 27, 2026), investor Dylan Koch said his AI bots scrape county records, read PDFs for mortgage balances, estimate ARV and drop records that don't pencil, then format the data, upload it to his CRM, skip trace and push records to a dialer. He also asked a bot to search the thousands of leads already in his CRM for active ones matching one specific buyer's criteria, and it surfaced two or three he had forgotten about that may turn into deals. The hosts are frank that bots still fail (one report broke when a bot connected to the wrong system), so Koch has a second 'chief of staff' bot review the others' output daily. The episode gives no deal results, only that the leads may become deals.
The prompt
1. Write down one active buyer's buy box: property types, areas or ZIP codes, price or max offer, beds/baths, condition they'll take on, and minimum equity or spread. Keep it about the property and the numbers only. 2. Export your seller leads from your CRM (address, property details, lead date, last contact, status, notes, any mortgage or equity info) and remove anyone marked do-not-contact. 3. Ask your AI: "Search these leads for properties that match this buy box. For each match, give the address, why it fits, what's unknown, and a rough spread using only the numbers in the data. Drop anything where what's owed means it can't work, and mark any estimate you're unsure of as [CHECK]. Don't invent values." 4. Then: "Rank the matches by how recently we spoke and how strong the fit is, and draft a short, friendly call-back note for each that mentions what they told us last time. No pressure or made-up deadlines." 5. Check each property's current details and ownership in public records yourself, confirm contact consent and do-not-call rules, then make the calls. Each week, have the AI (or a second check routine) list any lead it couldn't read or any step that failed, so errors don't pile up silently.
How to use it
- Copy the prompt above.
- Replace every [bracket] with your own details (or fill in the blanks on AgentPlaybook Real Estate, which remembers your answers for the next prompt).
- Paste it into ChatGPT, Claude, CRM and review the answer before you use it.
Works with: ChatGPT, Claude, CRM, county records, spreadsheet
Real-world example: collectingkeys.com/articles/how-investors-are-using-ai-agents-to-run-d
Featured in: AI prompts for writing SOPs, reports, policies and checklists
Before you use it: what to check
Already built into this prompt
- Don't invent values." 4.
- Each week, have the AI (or a second check routine) list any lead it couldn't read or any step that failed, so errors don't pile up silently.
- Check each property's current details and ownership in public records yourself, confirm contact consent and do-not-call rules, then make the calls.
Check yourself
- Check every figure, price and total against your own records before anything is sent, posted or booked. A person approves refunds, discounts, payments and price changes.
- Remove names, contact details, account numbers and other personal data the task doesn't need before you paste, and check your AI tool's data and training settings.
- Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details.
Using an AI agent? Download this prompt as a SKILL.md agent skill with these checks built in, or get the .zip for Claude.ai. How to install skills.
Want to make this prompt even better? Grade it with the free prompt grader or build your own with the AI prompt generator.
32 copy-paste agent prompts (PDF) plus 5 new prompts every week.
Get it free