MCP Server

Setup Guide

Connect Your AI to STR Data in 30 Seconds

One command is all it takes. Follow the setup guide for your AI client and start querying 20M+ Airbnb listings, market analytics, and revenue projections through natural language.

30-second setup

No coding required

Works with 5+ AI clients

Terminal
$ claude mcp add --transport http airroi \
  https://mcp.airroi.com \
  --header "X-API-KEY:your-airroi-api-key"

# That's it. 21 tools. 20M+ properties.
# Start asking questions in plain English.

30s

Setup Time

20

Tools Available

20M+

Properties

$10

Starting Credits

Before You Start

API Key

Sign up at AirROI, grab your sk_live_ key from the Developer Dashboard. $10 minimum, pay-as-you-go.

AI Client

Install Claude Code, Codex, Gemini CLI, Cursor, VS Code, or any MCP-compatible client.

Setup Instructions

Select your AI client below. Each tab walks through the complete setup process, from opening settings to verifying the connection.

Recommended

Claude Code is Anthropic's terminal-based AI coding agent. It supports MCP natively with a single CLI command -- no config files, no manual editing. This is the fastest path to connecting AirROI.

Step 1: Open your terminal

Launch your terminal application (Terminal on macOS, iTerm, Windows Terminal, or any shell). Claude Code must already be installed -- if not, run npm install -g @anthropic-ai/claude-code first.

Step 2: Run the setup command

This single command registers AirROI as an MCP server with HTTP transport and authenticates using your API key:

claude mcp add --transport http airroi https://mcp.airroi.com \
  --header "X-API-KEY:your-airroi-api-key"
Step 3: Verify the connection

Start a new Claude Code session and type /mcp to check the server status, or ask: “What MCP tools do you have from AirROI?” You can also verify with claude mcp list.

Replace your-airroi-api-key with your actual API key from the Developer Dashboard.

What to Ask After Connecting

Once connected, just ask your AI in plain English. The MCP server automatically selects the right tool and returns structured data.

Market Analysis

What's the average occupancy in Miami Beach for 2BR listings?

Compare ADR trends in Austin vs Nashville over the last 12 months

Investment Research

Find top 10 highest-revenue listings within 5 miles of downtown Denver

Estimate annual revenue for a 3BR property near Disneyland

Competitive Intelligence

How many active listings are in the Smoky Mountains?

What's the average lead time for bookings in Maui?

Portfolio Management

Get metrics for listing ID 12345678 including occupancy and revenue trends

What are future rates for comparable listings near my property?

Your First Query

Here is what a real interaction looks like after setup. You ask a question in natural language, your AI calls the appropriate tool, and returns structured market intelligence instantly.

claude — airroi-mcp-session
> What's the average occupancy rate in Miami Beach for 2BR listings?
Called search_markets("Miami Beach")
Called market_summary("Miami Beach", bedrooms=2)
  Miami Beach, FL — 2BR Performance

  Average Occupancy:  72.4%
  Average Daily Rate: $289
  RevPAR:             $209
  Active 2BR Listings: 3,847

  Seasonality:
    Peak (Dec-Mar):   89.1% occ · $342 ADR
    Shoulder (Apr-Jul): 68.5% occ · $278 ADR
    Low (Aug-Oct):    58.2% occ · $231 ADR

The market shows strong seasonal patterns with peak demand during winter months. At 72.4% average occupancy and $289 ADR, 2-bedroom listings in Miami Beach generate approximately $76,000 in annual revenue at the median. The summer shoulder season still holds above 68%, making this a relatively year-round market.

This is a representative example. Actual values depend on the current dataset and your query parameters.

Dive Deeper

Frequently Asked Questions

One command: claude mcp add --transport http airroi https://mcp.airroi.com --header "X-API-KEY:your-airroi-api-key". Replace the key with your actual API key from the Developer Dashboard. Claude can immediately query 20M+ Airbnb listings, market analytics, and revenue estimates through natural language.

No. If you can install Claude Desktop or Claude Code, you can use the MCP server. The one-command setup takes 30 seconds, and then you query data in plain English. No coding, no API integration, no data processing required.

Same pay-as-you-go pricing as the REST API. No subscriptions, no contracts, no monthly minimums. Buy API credits starting at $10, and each MCP tool call consumes the same credits as the equivalent REST API call. You only pay for the queries your AI actually makes.

Yes. Both the MCP server and the REST API use the same API key, the same credit balance, and return the same data. Use the MCP server for natural-language queries through Claude or Cursor, and the REST API for programmatic integrations in your own applications. They complement each other and share your pay-as-you-go credits.

When credits are exhausted, tool calls return a clear error message indicating insufficient credits -- no data loss, no silent failures. Your AI assistant will relay the message and suggest purchasing additional credits from the Developer Dashboard. You can top up instantly and resume querying without reconfiguring anything.

Any client that supports the Model Context Protocol (MCP) with HTTP transport. This includes Claude Code, OpenAI Codex, Google Gemini CLI, Cursor, VS Code with Copilot, Claude Desktop, Windsurf, and any MCP-compatible application. MCP is an open standard created by Anthropic and adopted by Google, OpenAI, and Microsoft, so compatibility is expanding rapidly.

Ready to Connect Your AI?

One command. 21 tools. 20M+ properties. Start querying in 30 seconds.
made with