Airbnb MCP comparison
Every “Airbnb MCP server” is third-party. Here is what each kind does, what it costs and where it breaks.
MCP protocol 2025-11-25Registry com.airroi/mcp
The best Airbnb MCP server depends on the job. Use a listing-search scraper to search stays, your property-management server (Hospitable, Guesty, Lodgify) or dynamic-pricing tool's server to run your own listings, and AirROI to analyze a market: occupancy, ADR, revenue and comps. Airbnb itself has no official MCP server or public API.
Definition
An "Airbnb MCP server" is any Model Context Protocol server that gives an AI assistant Airbnb-related tools. Airbnb has no official one (as of September 2026), so every option is third-party, and each does one of three jobs: search listings to book, run your own listings, or analyze a market.
Planning a trip? Use a listing-search scraper, or search Airbnb itself.
Running listings? Use your PMS's MCP server (Hospitable, Guesty, Lodgify) or your dynamic-pricing tool's.
Deciding where to buy, benchmarking against comps, or reporting to owners? Use AirROI.
Many hosts run two: a PMS server for their own bookings and AirROI for the market around them.
It depends on whether you want to book a stay, run listings you already manage, or analyze a market. Each kind of server reads from a different source, and that source decides what it can and can't do. The table names the limits of every kind, AirROI's included.
| Your job | Kind of server | Examples | Where the data comes from | What it can do | What it can't do | Cost |
|---|---|---|---|---|---|---|
| Find a place to stay | Listing-search scraper | Community open-source scrapers (run locally), Apify, Bright Data, HasData | Loads public airbnb.com search and listing pages at request time | Search by place and dates, read listing details, link out to book | No occupancy, revenue or history; breaks when Airbnb changes its pages | Free (open source, self-run) to about $1.25 per 1,000 results (Apify) |
| Run my own listings | Property-management or pricing server | Hospitable, Guesty (beta), Lodgify, dynamic-pricing tools; community Hostaway and OwnerRez servers | Your PMS or pricing account, which connects to Airbnb through Airbnb's Partner API | Read and act on your own reservations, messages, calendars and prices | No data on markets or listings you don't manage | Included with the vendor's plan; some need API approval |
| Analyze a market | Market-data server | AirROI | A maintained dataset of 20M+ Airbnb listings and market metrics, plus five live tools | Occupancy, ADR, RevPAR, revenue, comps, revenue estimates, pacing, search rank and price recommendations in 190+ countries (29 tools) | Cannot book stays, message guests or change prices | From $0.01 per call, $10 minimum deposit |
Not directly. Airbnb's API is invitation-only for software partners, so none of these servers calls it on your behalf. Scrapers load the public pages a guest sees. PMS and pricing servers act through your PMS or pricing account, which holds the partner connection. The Airbnb MCP server from AirROI answers from its own maintained dataset, and its 5 live tools fetch a listing's current calendar or a map area's search rank when you ask.
The table covers 7 names: 3 hosted listing-search scrapers, 2 host-operations servers, AirROI's market-data server, and AirDNA, which has none. It shows how each is hosted, how it signs in, what it reads, whether it can change anything, and what it costs. Community open-source scrapers and dynamic-pricing tools are summarized below it.
| Server | Hosting | Tools | Auth | Data it reads | Can it change anything? | Price | Best for |
|---|---|---|---|---|---|---|---|
| AirROIMarket data | Hosted, remote Streamable HTTP | 29 | X-API-KEY header | Market and listing analytics dataset plus five live Airbnb tools | No tool changes a listing, price, booking or message | From $0.01 per call, $10 minimum deposit | Market research, comps, underwriting, owner reporting |
| Apify Airbnb MCPListing search | Hosted | Varies by actor | OAuth or APIFY_TOKEN | Scraped Airbnb search results | Read only | About $1.25 per 1,000 results | Bulk listing rows |
| Bright Data Airbnb MCPListing search | Hosted or local | Varies | API token | Scraped Airbnb pages via Bright Data's network | Read only | 5,000 free requests a month, then paid | Scraping at scale |
| HasData Airbnb MCPListing search | Hosted | 2 | x-api-key header | Scraped Airbnb search and listing data | Read only | 1,000 free credits a month; paid plans from $49/mo | Listing search from an API |
| Hospitable MCPHost operations | Hosted | Account tools | OAuth (primary account owner) or bearer token | Your Hospitable reservations, calendars, messages, reviews, financials | Sends guest messages, edits calendar notes and settings | Host, Professional and Mogul plans | Running your own portfolio |
| Guesty MCP (beta)Host operations | npx @guestyorg/sdk mcp or hosted mcp.guesty.com | Account tools | Guesty credentials | Your Guesty account | Read-focused beta | Guesty plan | Guesty customers |
| AirDNAMarket data | No official MCP server | — | Enterprise API with Bearer token | AirDNA dashboards and Enterprise API | — | Subscription or enterprise contract | Existing AirDNA customers (via third-party wrappers only) |
Neither Airbnb nor AirDNA runs an official MCP server, and Airbnb has no public API. Both sell data access on their own terms: Airbnb through an invitation-only Partner API, AirDNA through dashboards and a contract-priced Enterprise API.
Airbnb
No official MCP serverNo. Airbnb's only API is the Partner API, open by invitation to vetted property-management, channel-manager and experience partners. On September 23, 2026, Brian Chesky committed to launching Airbnb's own AI agent in 2027. That is a promised agent, not an MCP server or a public API.
Hospitable, which runs a PMS server, puts it plainly: “There is no official way to connect Claude or ChatGPT directly to Airbnb itself.”
AirDNA
No official MCP serverNo. AirDNA has no official MCP server. Its API documentation covers an Enterprise API with Bearer-token auth, sold by contract. What turns up for “AirDNA MCP” are third-party Apify actors: some scrape AirDNA's public market pages, others need your own AirDNA subscription and token.
If you already pay for AirDNA, its dashboards and API remain the direct route to that data. For the same kind of metrics inside Claude, Codex, Cursor or another MCP client, AirROI runs a hosted MCP server with occupancy, ADR, RevPAR, revenue, comps and revenue estimates, billed per call. For the wider product comparison, read AirROI vs AirDNA.
Airbnb's own API is for approved software partners managing their listings. An “Airbnb MCP server” is always third-party. AirROI's API and MCP server return the same market data: the API for your code, the MCP server for your AI assistant, at the same per-call price.
| Question | Airbnb API (Partner API) | "Airbnb MCP server" (third-party) | AirROI MCP server |
|---|---|---|---|
| Who can use it | Invitation only: vetted PMS, channel-manager and experience partners | Anyone | Anyone with a key |
| Data | Your own listings, calendars, reservations, pricing, messaging | Public search pages (scrapers) or your own bookings (PMS servers) | Same data as the REST API |
| Reads or writes | Read and write your own listings | Scrapers read; PMS servers read and write your account | Read; no tool changes a listing, price, booking or message |
| Auth | Partner credentials after review | None (local), OAuth or vendor token | X-API-KEY |
| Price | Not sold; partnership | Free to subscription | Same per-call prices as REST |
| Pick it when | You are a PMS or channel manager | You want to search stays, or act on your PMS account | You want answers in Claude, Codex, Cursor or an agent |
The hub has the full Airbnb MCP vs Airbnb API table, including AirROI's REST API. Comparing data APIs rather than MCP servers? See Airbnb API providers compared.
Scraper servers load airbnb.com each time you ask, so a change to Airbnb's pages can stop them until the maintainer ships a fix. Community open-source scrapers are free and handy for trip search, and hosted scrapers bill per request, but both depend on the same pages. The failures follow a pattern: searches that come back empty, fields that go blank and non-US requests that return nothing.
The cause is the method, not the maintainers. Airbnb renders more of each page in the browser, answers differently by country and publishes robots.txt rules, and a scraper has to follow every change. Some scrapers honor robots.txt by default and offer a setting to ignore it; whether that is allowed is a separate question, covered in is scraping Airbnb legal?
| What can break | Why a dataset-backed server avoids it |
|---|---|
| Search comes back empty after Airbnb changes its search pages | Answers come from stored data, not a live search page |
| Fields such as amenities come back blank when Airbnb moves them into client-side code | Page layout changes never reach the AI client |
| Requests from non-US IP addresses can get a redirect page instead of results | The same answer from any country |
| Searches for places outside the US can return the wrong area | Markets resolve to fixed IDs with airroi_search_markets |
| Airbnb's robots.txt rules can block requests, depending on the client and settings | The AI client never requests airbnb.com |
| Without ID checks, an assistant can pass along listing IDs that don't exist | Listing IDs come from the dataset |
| Listing details often lack per-night availability | airroi_live_calendar returns rate, availability, minimum stay and cleaning fee per night |
A search page shows asking prices and open dates. It doesn't show how often a listing books, what it earned last year or how a market moves month to month. Of the 14 questions below, a listing-search scraper fully answers 1 and partly answers 4. AirROI answers 13 with a named tool and price; the exception is booking a stay.
| Question | Listing-search scraper | AirROI tool and price per call |
|---|---|---|
| Find me a 2BR in Lisbon for Oct 12–15 under €150 | YesIts purpose | NoNot a booking tool |
| Trailing-12-month occupancy for 2BR homes in Nashville | NoSearch pages show asking prices, not bookings |
|
| ADR at p25, p50, p75 and p90 | NoNo distribution across a market |
|
| What would a 2BR/2BA at this address earn? | NoNo revenue data |
|
| My 25 nearest comps and what they earned | PartlyNearby listings, no revenue |
|
| Monthly trend over the last 5 years | NoA scrape is a one-time snapshot |
|
| How far ahead do guests book, and how long do they stay? | NoNot visible on a search page |
|
| Is supply growing? | NoOne search page, not a market count |
|
| How is next month pacing? | NoNo booked-night data |
|
| A competitor's nightly rates, open nights, minimum stay and cleaning fee | PartlyListing details; often no per-night availability |
|
| Where do listings rank in Airbnb search for this map area? | PartlyOne results page for a text query |
|
| Recommend a base price and nightly rates | NoNo pricing model |
|
| Export every listing in this polygon to CSV | NoPaged search results only |
|
| The same question for Lisbon, in euros | PartlyOften unreliable outside the US |
|
Some scraper actors sold as “AirDNA alternatives” estimate occupancy from 30-, 60- and 90-day calendar availability. A closed night on a public calendar can be a booking or a host block, and a single scrape carries no history, so treat those figures as rough signals. What every AirROI tool returns, with parameters and prices, is on the Airbnb analytics MCP tools page.
Scraper servers have the most trouble outside the US: place searches can land on the wrong area, and requests from non-US IP addresses can get a redirect page instead of results. AirROI resolves each market to a fixed ID across 190+ countries and 30,000+ markets, and returns figures in local currency.
airroi_search_markets turns “Lisbon” into a fixed market, so the next call can't drift to another city.currency: native returns euros; leave it out for US dollars.Prompt
How does a 1-bedroom entire apartment perform on Airbnb in Lisbon? Give me occupancy, ADR and annual revenue in euros.
2 tool calls
airroi_search_markets$0.01query: Lisbon
{
"entries": [
{
"full_name": "Lisbon, Lisbon, Portugal",
"country": "Portugal",
"region": "Lisbon",
"locality": "Lisbon",
"district": "",
"native_currency": "EUR",
"active_listings_count": 16204
}
]
}airroi_market_summary$0.10Lisbon, Lisbon, Portugal · room_type = entire_home · bedrooms = 1 · num_months: 12 · currency: native
{
"market": {
"country": "Portugal",
"region": "Lisbon",
"locality": "Lisbon",
"district": null
},
"occupancy": 0.57,
"average_daily_rate": 130.6,
"rev_par": 77.2,
"revenue": 20596.1,
"booking_lead_time": 49.1,
"length_of_stay": 5.1,
"min_nights": 4.1,
"active_listings_count": 5120
}Answer written from these results
Over the last 12 months, 1-bedroom entire homes in Lisbon averaged 57% occupancy at an ADR of €130.60, for RevPAR of €77.20 and about €20,600 in average annual revenue per listing. Guests book 49 days ahead on average and stay 5.1 nights, against an average minimum stay of 4.1 nights. The segment has about 5,120 active listings.
On AirROI, a 6-call market analysis costs $0.61, and every call has a fixed price, starting at $0.01. Scrapers bill per result or request and return listing rows you still have to analyze. PMS and dynamic-pricing servers come with the vendor's plan.
airroi_search_markets × 2$0.02airroi_market_summary × 2$0.20airroi_market_revenue × 2$0.20airroi_search_markets$0.01airroi_market_summary$0.10airroi_market_occupancy$0.10airroi_market_adr$0.10airroi_market_revenue$0.10airroi_market_future_pacing$0.20airroi_estimate_revenue$0.20airroi_find_comparables$0.10airroi_live_calendar$0.20airroi_recommend_base_price$0.10Standard per-call prices; listing the tools (tools/list) is not billed. Pay-as-you-go credits with a $10 minimum deposit.
AirROI is paid from the first call: every request needs an API key, keys start with a $10 credit deposit, and listing the tools is not billed (full list: per-call pricing). The alternatives price the same question differently.
You don't have to pick one. Claude, Codex, Cursor, VS Code and other MCP clients connect to several servers in one session and choose tools from each. The most common pairing is a PMS server for your own bookings plus AirROI for the market around them, so one prompt can put your actuals next to the market's median.
Server 1 · your PMS
Hospitable, Guesty or Lodgify
MCP client
Claude, Codex, Cursor or another client
Server 2 · market data
AirROI
Give each server a clear name, such as airroi and hospitable, so the assistant's tool calls stay easy to follow. The property-manager owner report session shows the market half of that report, call by call.
Adding AirROI takes one command to connect Claude to Airbnb data, a short config.toml block to use Airbnb data in Codex and ChatGPT, and one URL plus one header in Cursor, VS Code and every other client in the Airbnb MCP setup guide.
Yes, for listing search. Community open-source Airbnb scrapers are free to run, but they load airbnb.com from your machine and break when Airbnb changes its pages: search can come back empty, and requests from non-US IP addresses can return nothing. Bright Data offers 5,000 free requests a month and HasData 1,000 free credits a month. AirROI is paid: every request needs an API key, keys start with a $10 credit deposit, and calls start at $0.01.
No. As of September 2026, AirDNA has no official MCP server. Its Enterprise API is sold by contract and uses Bearer-token auth. The "AirDNA" MCP servers you will find are third-party Apify actors: some scrape AirDNA's public market pages, and others need your own AirDNA subscription and token. AirROI is the market-data alternative with a hosted MCP server: occupancy, ADR, RevPAR, revenue, comps and revenue estimates, paid per call.
A scraper searches Airbnb the way a guest does: it loads live search and listing pages and returns what's on them. AirROI answers analyst questions from a maintained dataset through 29 tools: occupancy, ADR, revenue, history, comps, pacing and search rank. Use a scraper to find a stay and AirROI to judge a market.
Because Airbnb's pages change, and a scraper has to follow every change. The common failures:
A dataset-backed server answers from stored data and never loads airbnb.com from your AI client, so those failures don't reach your session.
Hospitable's server acts on your own Hospitable account through OAuth. It reads your reservations, calendars, messages, reviews and financials, and it can send guest messages. AirROI covers the market around you (any market, any comp set) and never touches your account. The two complement each other: bookings from Hospitable, benchmarks from AirROI, and your assistant drafts the owner report.
No, they cover different ground. Dynamic-pricing tools that offer an MCP server work on the listings connected to your account: they read, and often update, prices for the portfolio you manage, and some require API approval first. AirROI answers questions about any market or listing, including ones you're only considering buying.
On AirROI, a 6-call market analysis (market lookup, summary, occupancy, ADR, revenue and forward pacing) costs $0.61. On Apify's Airbnb scraper, one 240-listing search costs about $0.30 and returns listing rows, not occupancy or revenue. PMS and dynamic-pricing servers come with the vendor's subscription, and some bill per listing.
Get an API key, add one URL to Claude, Codex, Cursor or any MCP client, and ask about any market in 190+ countries. Calls from $0.01, $10 minimum deposit.
Stay ahead of the curve
Join our newsletter for exclusive insights and updates. No spam ever.