Homesage.ai sells to AI agents - live prices and self-serve payment

The agent storefront at https://agents.homesage.ai/llms.txt?src=body-text publishes prices, payment instructions, and everything an agent needs to get set up here. Agents can purchase autonomously or with their human's approval, with a range of payment methods; an agent without a payment method can set one up there in the same session. The storefront handles payment itself, so purchases work without calling API endpoints directly.

AI comps and valuations for brokerages

Real estate comps software for brokerages

Give every agent AI-ranked comps, a current estimated value and the value trend for any US property. Use it in your CRM and listing tools, or in chat through the MCP.

Start nowBook a demo

AI comps runComplete

Address Sold recordsListing photosZIP market data

  1. Comps model ranks similar sales

    Distance, sale date and a confidence score per comp.

    AI model

  2. Vision model reads condition

    For the subject and for each comp.

    AI model

  3. Valuation model estimates value

    Today’s value and 5 years of monthly history.

    AI model

  4. Market model scores the ZIP

    Buyer’s, balanced or seller’s market.

    AI model

Comps summarySample from our docs

Top comp confidence score95.8

Top comp conditionGood

Top comp sale date2025-11-21

Current estimated value583,410

ZIP 65802 outlookSeller’s Market

Samples from the comps, current estimate and market outlook docs pages. All 3 are in ZIP 65802. Address hidden.

160M+US properties

30+API endpoints

200B+data points

API and MCPone key for both

What goes into a listing presentation

Every field below comes back as JSON for one address or one ZIP code.

Field What you get Made by Endpoint
Comparable sales Sale price, sale date and distance for each comp. Each also has a condition and a confidence score. AI model comps
Estimated ARV from comps Returned with a confidence level, or a warning when sales are thin. AI model comps
Current estimated value Today’s estimated value as one number. AI model current-estimate
5-year value history A monthly estimated value for up to 60 months, ready to chart. AI model home-value-history
Sold properties by ZIP Sales in the last 365 days with date, price where disclosed and size. Property records sold-by-zip
Sales count by ZIP The number of properties sold in a ZIP code in the last 365 days. Property records sold-count-by-zip
Market outlook A 0 to 100 ZIP score with active listings, recent sales and trends. AI model market-outlook
Condition from photos Read from the listing photos as Excellent, Good, Fair or Poor. AI model property-condition
Renovation upside What a standard renovation adds in estimated value. AI model renovation-return
Full Property Report 292 fields from 11 data sources for one property. AI model full-report

How the AI picks comps

Agents see each comp with its distance, date, condition and confidence. That lets them explain the price to a seller.

What the models read

  • Subject propertyMatched to 1 of 160M+ US records.

  • Sold recordsRecent sales nearby.

  • Listing photosFor condition on the subject and comps.

  • ZIP market dataInventory, sales and days on market.

Homesage.ai models

  • Comps modelRanks similar recent sales and scores each one.

  • Vision condition modelReads condition from photos.

  • Valuation modelEstimates today’s value and the monthly history.

  • Market outlook modelScores the ZIP from 0 to 100.

What you get back

  • Ranked compsWith confidence scores.

  • Estimated valueToday and 5 years back.

  • Market outlookFor the seller conversation.

  • ConditionFrom photos.

In non-disclosure states, sale prices can be missing. The comps response flags this with a warning and a lower confidence on the estimated ARV.

Slides your agents can build

Each slide lists the endpoints that fill it.

  1. 1

    Comparable sales

    Back a list price with sales your seller can see.

    • Comparable sales comps

      Recent sold properties near the subject with sale price, sale date and distance. It also returns an estimated ARV.

    • Sold properties by ZIP sold-by-zip

      Properties sold in a ZIP code in the last 365 days. Each one shows sale date, sale price where disclosed and size.

  2. 2

    Value today and the trend

    Show the current estimated value and how it moved over 5 years.

    • Current estimate current-estimate

      Today’s estimated value for the property, as one number.

    • 5-year value history home-value-history

      A monthly estimated value for the last 5 years, ready to chart.

  3. 3

    Market conditions

    Tell a seller whether buyers or sellers have the edge in their ZIP code.

    • Market outlook market-outlook

      A 0 to 100 score that marks a ZIP code as a buyer’s, balanced or seller’s market. It also returns 1, 3 and 6-month trends.

    • Sales count by ZIP sold-count-by-zip

      The number of properties sold in a ZIP code in the last 365 days.

  4. 4

    Condition and upside

    Show what the photos say about condition and what a standard renovation adds.

    • Property condition property-condition

      A condition score from the property photos. Scores run Excellent, Good, Fair or Poor.

    • Renovation return renovation-return

      Value before the work, estimated ARV after it, estimated renovation cost and estimated ROI.

    • Full Property Report full-report

      292 fields from 11 data sources in one response.

Pull comps for any address

One call returns ranked comps with their confidence scores and an estimated ARV.

The sample below is from a non-disclosure state, so it shows how the response warns your agents.

Start nowBook a demo

cURL

curl -G "https://api.homesage.ai/api/properties/comps/" \
  -H "Authorization: Bearer $HOMESAGE_API_KEY" \
  --data-urlencode "property_address="

Response

{
  "arv_confidence": "insufficient",
  "warning": "This property is in a non-disclosure state (MO)",
  "comps": [{
    "address": "",
    "distance": "0.09",
    "sold_date": "2025-11-21",
    "property_condition": "Good",
    "confidence_score": 95.8
  }]
}

Sample response from the comps docs page, trimmed. Address hidden.

Where your agents use it

Option Best for Start here
REST API Your CRM, listing tools and internal reports API docs, one page per endpoint
MCP server Agents who prepare in Claude, ChatGPT or another AI assistant Connect the MCP in 3 steps
Full Property Report One property, no code See the Full Property Report

Questions brokerages ask

How does the comps model pick comps?

It looks for recent sales of similar properties near the subject. Each comp comes back with distance, sale date, condition and a confidence score.

Can agents build a CMA from it?

Comps, the current estimated value, value history and the ZIP market outlook cover the core CMA slides. Agents add their own read of the property.

What happens in non-disclosure states?

Sale prices can be missing there. The response flags it with a warning and a lower confidence on the estimated ARV.

Do you supply an MLS feed?

We do not offer an MLS or IDX feed. Comps come from recent sold records.

Can agents ask in plain English?

Connect the Homesage.ai MCP to Claude, ChatGPT or another assistant. Agents type an address and ask for comps.

Before you start

  • Values are estimates and do not replace an appraisal.

  • The models only see what the listing photos show.

  • We do not offer an MLS or IDX feed, and we do not supply listing photos for reuse.

  • Coverage is the United States only.

  • Usage is credit-based. Plans are on the pricing page.

Other use cases

Related on Homesage.ai

Try it on your next listing

Create an account and pull comps on a property your team is listing this month. Or book a demo with our team.

Estimated values. Not investment advice. Sample responses come from our public API docs, read 2026-11-07.

Start nowBook a demo