---
name: al-388-write-a-cma-with-ai-without-giving-it-your-mls-login-you
description: "Many agents want AI help with comparative market analyses but don't want an AI agent logging into the MLS. On Oct 28, 2026 David Ursino, a managing broker who teaches AI to Canadian real estate agents, posted a demo of a CMA where the AI never touches the MLS: he does the search and export himself, Claude Code reads the exported folder, builds the adjustment table, asks him for his adjustments, and then writes the CMA. The prompt below follows the same split, so the judgement calls on value stay with you. Use when the user wants help with comparative market analysis (sales) for a small business."
---

# Write a CMA with AI without giving it your MLS login: you search and export the comparables, Claude reads the folder, builds the adjustment table, asks you for each adjustment, then writes the CMA

Many agents want AI help with comparative market analyses but don't want an AI agent logging into the MLS. On Oct 28, 2026 David Ursino, a managing broker who teaches AI to Canadian real estate agents, posted a demo of a CMA where the AI never touches the MLS: he does the search and export himself, Claude Code reads the exported folder, builds the adjustment table, asks him for his adjustments, and then writes the CMA. The prompt below follows the same split, so the judgement calls on value stay with you.

## Inputs to collect first
Ask the user for anything below that they haven't already given you. Don't guess missing facts.

- Ress, property type, beds, baths, square footage, lot size, year…
- Or paste the MLS export of sold, active and expired comparables

## Instructions

Do the task below with the user's inputs in place of each [bracket].

Help me write a comparative market analysis (CMA) from comparables I've already exported myself. You won't log into the MLS or look anything up; use only the files I give you. Subject property: [address, property type, beds, baths, square footage, lot size, year built, condition, key features]. Comparables: [attach or paste the MLS export of sold, active and expired comparables]. 1) Read the files and list each comparable in a table with sale or list price, date, distance from the subject, size, beds/baths, lot, age, condition and notable features. Mark any comparable that looks like a poor match and say why. 2) Build an adjustment table showing each difference from the subject property (size, bedrooms, bathrooms, garage, lot, condition, updates, location). Leave every adjustment amount blank and ask me for my figure for each type of difference before you calculate anything. 3) Once I give you my adjustments, work out each adjusted price, show the math, and suggest a price range with one plain sentence on how you got there. 4) Write a short CMA summary for the seller in plain English: the market right now, the three most relevant comparables, the suggested range, and what could move the price up or down. Never invent a comparable or a figure; if something is missing from the export, write 'not in the data'.

## Safety checks (share these with the user)

- 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.
- Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details.
- Say clearly which parts need a person to review or approve before anything is sent, paid, posted or filed.

## About this skill

- Works with: Claude Code, Claude, ChatGPT, your MLS export, Google Sheets, Excel
- Based on this real-world example: https://x.com/david_ursino/status/2115383137249198177
- Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-388-write-a-cma-with-ai-without-giving-it-your-mls-login-you-sea/
- To install in an agent that supports Agent Skills, save this file as SKILL.md in a folder with the same name as `name` above (for example in your skills directory). Install steps for Claude.ai, Claude Code, Cursor and Codex: https://agentic-library.netlify.app/skills/