AirROI Airbnb MCP Server

Short-term rental market data for AI assistants · Remote · Streamable HTTP · com.airroi/mcp

AirROI's Airbnb MCP server gives Claude, Codex, Cursor, VS Code and any other Model Context Protocol client 29 tools for Airbnb and short-term rental market data: occupancy, ADR, RevPAR, revenue percentiles, comparables, revenue estimates, live Airbnb calendars and search rank, and price recommendations. It covers 20M+ listings in 190+ countries and answers from a maintained dataset, so your assistant never scrapes airbnb.com.

This repository is documentation only. The server is fully remote at https://mcp.airroi.com; there is nothing to install or run. AirROI is an independent data provider and is not affiliated with Airbnb. As of September 2026, Airbnb has no official MCP server or public API.

Which Airbnb MCP server do you need?

"Airbnb MCP server" can mean three different things. Pick by the job:

A side-by-side comparison with pricing and limitations: https://www.airroi.com/mcp-server/compare

Get an API key

1. Sign up for a free API key at https://www.airroi.com/api
2. Activate it in the developer dashboard at https://www.airroi.com/api/developer/activate and copy the key.
3. Use it as the X-API-KEY header value in the snippets below. Replace only YOUR_API_KEY.

Discovering tools (tools/list) is not billed, but every request needs the key.

Setup

Full, per-client guides: Claude · Codex and ChatGPT · every other client

Claude Code

claude mcp add --transport http airroi \
  https://mcp.airroi.com \
  --header "X-API-KEY: YOUR_API_KEY" --scope user

Check it with claude mcp get airroi, then ask Claude a market question.

Claude Desktop

Claude Desktop connects to remote servers with custom headers through the mcp-remote bridge (needs Node.js). Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows), then restart the app:

{
  "mcpServers": {
    "airroi": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mcp.airroi.com",
        "--header",
        "X-API-KEY:${AIRROI_API_KEY}"
      ],
      "env": { "AIRROI_API_KEY": "YOUR_API_KEY" }
    }
  }
}

Keep X-API-KEY:${AIRROI_API_KEY} free of spaces around the colon.

Codex (CLI, IDE extension and ChatGPT desktop app)

All three read ~/.codex/config.toml:

[mcp_servers.airroi]
url = "https://mcp.airroi.com"
env_http_headers = { "X-API-KEY" = "AIRROI_API_KEY" }

Then export AIRROI_API_KEY=YOUR_API_KEY in your shell profile. codex mcp list shows Auth: Unsupported for header auth; that is expected. ChatGPT on the web (developer mode) doesn't support API-key servers, so use Codex.

Cursor

~/.cursor/mcp.json (global) or .cursor/mcp.json (project):

{
  "mcpServers": {
    "airroi": {
      "url": "https://mcp.airroi.com",
      "headers": { "X-API-KEY": "${env:AIRROI_API_KEY}" }
    }
  }
}

VS Code (Copilot agent mode)

.vscode/mcp.json. The top-level key is servers, not mcpServers:

{
  "inputs": [
    {
      "type": "promptString",
      "id": "airroi-key",
      "description": "AirROI API key",
      "password": true
    }
  ],
  "servers": {
    "airroi": {
      "type": "http",
      "url": "https://mcp.airroi.com",
      "headers": { "X-API-KEY": "${input:airroi-key}" }
    }
  }
}

Devin Desktop (formerly Windsurf)

~/.config/devin/mcp_config.json (Windows: %APPDATA%\devin\mcp_config.json):

{
  "mcpServers": {
    "airroi": {
      "serverUrl": "https://mcp.airroi.com",
      "headers": { "X-API-KEY": "${env:AIRROI_API_KEY}" }
    }
  }
}

Antigravity CLI (Gemini CLI's successor)

~/.gemini/config/mcp_config.json (global) or .agents/mcp_config.json (project):

{
  "mcpServers": {
    "airroi": {
      "serverUrl": "https://mcp.airroi.com",
      "headers": { "X-API-KEY": "YOUR_API_KEY" }
    }
  }
}

n8n

1. Add an AI Agent node, then attach an MCP Client Tool node to it.
2. Set the endpoint to https://mcp.airroi.com and the transport to HTTP Streamable.
3. Choose Header Auth and create a credential with name X-API-KEY and your key as the value.

OpenAI Agents SDK (Python)

import asyncio, os
from agents import Agent, Runner
from agents.mcp import MCPServerStreamableHttp

async def main():
    async with MCPServerStreamableHttp(
        name="airroi",
        params={
            "url": "https://mcp.airroi.com",
            "headers": {"X-API-KEY": os.environ["AIRROI_API_KEY"]},
        },
    ) as airroi:
        agent = Agent(name="STR analyst", mcp_servers=[airroi])
        result = await Runner.run(
            agent, "Occupancy and ADR for 2-bedroom homes in Nashville?"
        )
        print(result.final_output)

asyncio.run(main())

Any stdio-only client

npx mcp-remote https://mcp.airroi.com --header X-API-KEY:${AIRROI_API_KEY}

A real session

Prompt, run against the live server at mcp.airroi.com:

Compare Austin and Nashville for a 2-bedroom entire-home Airbnb: occupancy, ADR, RevPAR, annual revenue, and how many competing listings I'd face.

Tool calls: airroi_search_markets → airroi_search_markets → airroi_market_summary → airroi_market_summary → airroi_market_revenue → airroi_market_revenue (cost $0.42)

Over the trailing 12 months, 2-bedroom entire homes in Nashville earned more per listing: $40,075 in average annual revenue versus $24,965 in Austin. Occupancy is nearly identical (49% vs 48%), so the gap comes from rate: Nashville's ADR is $296 against Austin's $236, for RevPAR of $146 versus $114. Nashville also has less 2-bedroom supply, about 1,270 active listings a month versus about 2,029 in Austin. October 2025 was the strongest month for the median listing in both markets, at $5,009 in Nashville and $3,941 in Austin.

More sessions (live search rank, owner reports, forward pacing, Lisbon): https://www.airroi.com/mcp-server/use-cases

Tools (29)

No tool can change a listing, price, booking or message. 26 tools are flagged read-only; the other 3 create a download file or a scan cursor.

Market analytics (11)