The real estate MCP server

Subdivvy is a real estate MCP server. Point your AI at it and get clean, structured property data: ownership, values, sales, permits, and distress signals. Ask a question, get a real answer.

Start free trial Read the API docs →

What is an MCP, in plain English?

MCP is an open standard that plugs an AI assistant into outside tools and data. Think of it as a universal adapter: build one connector, and any MCP client can use it.

Subdivvy speaks MCP, so you connect once and your assistant does the rest:

1.Connect Subdivvy with a URL or a config block.
2.Ask about a property, an owner, or a market.
3.Get an answer from live data. No glue code.

Why organized property data matters

Your AI is only as good as the data it can reach. Public records sit on thousands of county sites in a dozen formats, and no model can reason over that mess.

Subdivvy aggregates it, cleans it, and scores it, then serves one clean structure. You get answers, not guesses.

See it connect

Two ways to connect. Both take about two minutes.

Soon

Connect via custom connector. For Claude and other desktop apps: paste the URL and sign in.

Soon

Connect via AI terminal. For Codex, Cursor, and CLI clients: add the config block.

Connect it yourself

Paste the URL into a remote connector, or drop this block into a config file:

{
  "mcpServers": {
    "subdivvy": {
      "type": "http",
      "url": "https://mcp.subdivvy.com/api/mcp"
    }
  }
}

Endpoint mcp.subdivvy.com/api/mcp. Sign in with OAuth or use an API key. Step-by-step guides for Claude, ChatGPT, and Cursor.

Tools your AI can call

find_motivated_sellers

Ranks likely-to-sell owners in an area by a 0–22 motivation score.

score_property

Turns one address into a motivation score with a signal-by-signal breakdown.

property_profile

Returns a full record: owner, values, sale history, and distress signals.

comp_analysis

Pulls recent comparable sales near an address from public records.

rental_yield

Estimates gross rental yield for an address or ZIP.

mao_calculator

Computes the maximum allowable offer from ARV, repairs, and target profit.

Example query

› Find high-equity absentee owners in Gainesville and score the top three.

→ A ranked list with owner, equity, and a motivation score for each, live from Subdivvy.

Questions

What is a real estate MCP?

It stands for Model Context Protocol. It lets AI assistants like Claude read and query a real estate database directly, in real time, instead of guessing from training data.

Which AI tools support MCP?

Any MCP-compatible client, including Claude and Cursor, plus a direct REST API for everything else.

How is the property data updated?

Subdivvy refreshes its datasets as new county recorder and public tax data comes in.

Start free trial Browse integrations →