Model Context Protocol

Connect AI Agents to 160M+ US Parcels

Give Claude, GPT, and AI agents instant access to parcel data. Query properties by address, APN, or location — all through natural language.

View Setup Guide

What is MCP?

The Model Context Protocol (MCP) is an open standard developed by Anthropic that enables AI systems to connect to external data sources and tools. Think of it as a universal adapter that lets AI agents discover and use capabilities without custom integration code.

For parcel data, MCP eliminates the friction of reading API docs, writing HTTP clients, and handling authentication. Your AI agent sees available tools, understands what they do, and calls them as needed during conversations.

Open standard — works with Claude, GPT, and more

Example Conversation

User: "I'm looking at a property at 550 South Hope 
Street in Los Angeles. What can you tell me about it?"

Claude: "I'll look up that property for you."

[Claude calls lookup_parcel_by_address]

"Here are the details for 550 S Hope St, Los Angeles, CA:

• APN: 5139-017-020
• Parcel Size: 1.24 acres
• Land Use: Commercial Office
• Year Built: 1983
• Total Value: $127.5M
• Owner: Brookfield Properties LP
• Zoning: C2-1 Commercial"

Why Use MCP for Parcel Data?

Traditional API integration takes hours. MCP takes minutes.

AI-Native Access

AI agents discover and call parcel data tools automatically. No API documentation to read, no authentication code to write.

Natural Language Queries

Ask "Who owns 550 S Hope St in Los Angeles?" and get parcel details, ownership, and value data instantly.

Multi-Tool Chains

AI agents can query parcel data, then use results to call other tools — analyze, visualize, or enrich automatically.

Zero Integration Code

Install the MCP server once. AI handles the rest: parameter mapping, error handling, and response formatting.

Available Parcel Data Tools

AI agents can call these tools directly based on your natural language requests.

lookup_parcel_by_address

Query parcel data by street address

"Find the parcel at 123 Main St, Austin, TX"

lookup_parcel_by_apn

Query by Assessor Parcel Number (APN)

"Get details for APN 5139-017-020 in Los Angeles County"

search_parcels_by_location

Find parcels within radius of coordinates

"Show me parcels within 500m of 30.2672, -97.7431"

get_parcel_ownership

Get owner name and mailing address

"Who owns the property at this parcel ID?"

get_parcel_value

Get assessed and market value data

"What is the assessed value of this property?"

Quick Setup

Three steps to connect AI agents to parcel data.

1

Get an API Key

Sign up at getparceldata.com and request a free sample or purchase a data package. Your API key will be available in your account dashboard.

2

Install the MCP Server

Download and configure the GetParcelData MCP server with your API key.

{
  "mcpServers": {
    "getparceldata": {
      "command": "node",
      "args": ["/path/to/getparceldata-mcp-server.js"],
      "env": {
        "GETPARCELDATA_API_KEY": "your_api_key_here"
      }
    }
  }
}
3

Start Querying

Open Claude Desktop and start asking questions about properties. The AI will automatically use the MCP tools to retrieve parcel data.

Use Cases

How teams are using parcel data with AI agents.

Real Estate Investment Analysis

AI agents can research properties, pull comparable sales data, and generate investment summaries automatically.

Solar Site Assessment

Query parcel boundaries, ownership, and zoning to pre-qualify solar installation sites without manual research.

Insurance Underwriting

Instant property details for risk assessment: construction year, square footage, land use, and flood zone data.

Location Intelligence

Retail expansion teams query parcel data by geography to analyze competition, demographics, and site viability.

160M+ Parcels, One Connection

The GetParcelData MCP server provides access to our complete US parcel database — 160+ million parcels across 3,000+ counties, normalized into a unified schema.

160M+
US Parcels
3,000+
Counties
50
States
95%+
Coverage
Explore Coverage Map

Frequently Asked Questions

Everything you need to know about using parcel data with AI agents.

What is MCP and how does it work with parcel data?
MCP (Model Context Protocol) is an open standard that lets AI systems connect to external data sources and tools. With the GetParcelData MCP server, Claude, GPT, and other AI agents can query parcel data directly — discovering available tools, understanding parameters, and retrieving property information as part of natural conversations.
Which AI platforms support MCP?
MCP is supported by Claude Desktop, OpenAI GPTs with function calling, and other AI agents that implement the protocol. We provide configuration examples for Claude Desktop and are expanding support for additional platforms.
Do I need to write code to use the MCP server?
No. Once you install the MCP server and configure it with your API key, AI agents handle all the integration. You interact with parcel data through natural language conversations — the AI translates your requests into API calls and presents the results.
What data is available through the MCP server?
The MCP server provides access to all 160M+ US parcels in the GetParcelData database: property addresses, APNs, ownership information, assessed values, parcel boundaries (geometry), land use codes, zoning, building details, and more.
Is there a cost to use the MCP server?
The MCP server is free to install and use. You only pay for the parcel data queries you make through the GetParcelData API, with the same transparent pricing as our standard API access. Individual counties start at $49, or use our pay-as-you-go option for lower volume needs.
How do I get started?
1) Sign up at getparceldata.com for an API key. 2) Install the MCP server via npm. 3) Add it to your Claude Desktop configuration. 4) Start asking questions about properties. Full documentation and code examples are included with the server.

Ready to Connect AI Agents to Parcel Data?

Get started with a free API key and connect your AI agents to 160M+ US parcels in minutes.

View API Documentation