# Airbnb MCP

> Airbnb stays worldwide: search by place, dates, guests with total prices and links; listing details.

- name: ai.wondel/airbnb
- version: 1.0.0
- connection class: R0 (autonomous)
- trust: 83/100 flags: no-repository
- quality: 81/100 (new)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/ai.wondel%2Fairbnb
- tools: https://api.protogrid.dev/v1/servers/ai.wondel%2Fairbnb/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "airbnb": {
      "type": "http",
      "url": "https://mcp.wondel.ai/airbnb/mcp"
    }
  }
}
```


## Trust

- hygiene: 80
- liveness: 100
- freshness: 100
- provenance: 55
- drivers: +dns-namespace +website +reachable +uptime-100% +updated-5d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 81/100 (new)

### protocol: 100 (weight 25)

- pass `protocol.modern`: supports 2026-07-28 (server/discover)
- pass `protocol.stateless`: answers without a session
- pass `protocol.transport`: streamable HTTP
- pass `protocol.list_ttl`: tool list is cacheable (ttlMs 3600000)

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 64 (weight 30)

- pass `tools.descriptions`: every tool has a description
- warn `tools.description_length`: 2 tools have descriptions over 1,024 characters, which crowds the context
- pass `tools.schemas`: every tool has an object input schema
- fail `tools.annotations`: no tool declares readOnlyHint or destructiveHint, so clients cannot tell safe calls from risky ones
- fail `tools.directory_hints`: no tool declares all four of readOnlyHint, destructiveHint, idempotentHint and openWorldHint; missing: airbnb_listing (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); airbnb_search (readOnlyHint, destructiveHint, idempotentHint, openWorldHint)
- pass `tools.token_cost`: about 1,925 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: n/a (weight 15)

- n/a `stability.changes`, `stability.rug_pull`: tool history recorded for 5 of the 7 days needed

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 2, about 1,925 tokens to load them all
- tool set hash: 9cc46772e3e51776b0b2bb3039e402e4
- badge: [![protogrid quality](https://protogrid.dev/badge/ai.wondel/airbnb.svg)](https://protogrid.dev/servers/ai.wondel/airbnb)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

_No tool definition changes recorded._

## Remotes

- https://mcp.wondel.ai/airbnb/mcp (streamable-http, auth none, reachable true, uptime30d 1)

## Packages

_none_

## Tools (2)

- `airbnb_listing`: Full details of one Airbnb listing: name, type, capacity, bedrooms/beds/baths, sleeping arrangements, amenities (and notable ones missing), description (trimmed), house rules with check-in/check-out times, the host (name, Superhost, company or individual, years hosting, response rate — no contact details), the rating with its categories, approximate location, highlights and photos. With checkin/checkout it also returns the price for that stay and guests — the TOTAL incl. all fees (`price`), `price_per_night`, `price_before_discount`, `price_breakdown` — and `available`: false when the listing cannot be booked for those dates or that many guests (the note says why). Without dates there is no price (Airbnb prices only dated stays). Scraped live from airbnb.pl (typically ~1 s via direct fetch; ~10 s when it falls back to an unblocker). Two requests with dates (the page, then a dated search around it for the price). Result: { source, fetchedAt, query, page_url, listing: { id, url, name, kind, room_type, property_type, summary, overview, capacity, bedrooms, beds, baths, location: { name, lat, lng, approximate }, rating, review_count, rating_categories, guest_favorite, is_new, host: { name, superhost, verified, business, years_hosting, rating, review_count, response_rate, response_time, about }, description, checkin_from, checkin_until, checkout_by, house_rules, additional_rules, sleeping: [{ room, beds }], amenities, amenities_not_included, highlights, photos, stay, guests, available, free_cancellation, price_basis, price, price_per_night, price_per_month, price_before_discount, nights, currency, price_text, price_breakdown }, note? }. Text fields are Airbnb's Polish wording.
- `airbnb_search`: Search places to stay on Airbnb — apartments, houses, rooms, cabins, villas and hotels — in any city, region or country, for given dates and guests: price for the stay, rating and review count, Guest favourite / Superhost badges, beds and baths, coordinates, photos and a link that opens the same stay. Filter by room type, minimum bedrooms and total price. One page is up to 18 stays; `page` 2–15 for more (Airbnb caps a search at 270). Pass a listing's `id` to airbnb_listing for amenities, house rules and the host. Scraped live from airbnb.pl (typically ~1 s via direct fetch; ~10 s when it falls back to an unblocker). WITH checkin + checkout, `price` is the TOTAL for the whole stay for the given guests, all fees and taxes included (Airbnb's EU total-price display), in `currency`; `price_per_night` = price ÷ nights; `price_before_discount` is the struck-through total when discounted; `price_breakdown` lists Airbnb's own lines (e.g. '2 noce x 245,00 zł' → 490, 'Podatki' → 8). Stays of 28+ nights are quoted per month: `price` null, `price_per_month` set. WITHOUT dates, Airbnb prices each listing for its own suggested stay (its `stay`, usually ~5 nights soon), so prices are not comparable and not a quote — a `note` says so; pass dates to quote a trip. Only stays bookable for the given dates are listed. Result shape: { source, fetchedAt, query, search_url, place: { query, place_id, resolved, city, region, country }, stay: { checkin, checkout, nights } | null, guests: { adults, children, infants, pets }, currency, total_count, page, total_pages, count, listings: [...], note? } — the list key is `listings`; `total_count` is Airbnb's count of matches (more than it lets you page through). Each listing: { id, url, name, kind, city, details, bedrooms, beds, baths, host_type, rating, review_count, is_new, guest_favorite, superhost, badges, free_cancellation, stay, price_basis, price, price_per_night, price_per_month, price_before_discount, nights, currency, price_text, price_breakdown, lat, lng, photos }. `rating` is 0–5; `kind`/`city`/`details`/`price_text` are Airbnb's Polish wording (e.g. kind 'Mieszkanie' = apartment, 'Pokój' = room, 'Dom' = house); `host_type` is business | private; coordinates are approximate (Airbnb blurs them). An empty match answers count: 0 with a `note` saying why and what to change — that is a successful result, not an error.