LIVE SCRAPER EXPERIMENTAL

Booking.com

Booking.com’s search results page, read for your agent: for a location and a stay window, each property’s name, nightly price, review score and count, star class, image and a direct booking link. It’s experimental for an honest reason — Booking sits behind AWS WAF, which challenges datacenter traffic, so every call goes through an unblocker chain and can still come back blocked. When that happens, google_hotels_search is the reliable route, and it lists Booking’s prices among other sites anyway.

1 tool ~3–20 s cache: 5 min Any ISO code Booking.com search page via unblocker chain (AWS WAF)

Thirty seconds, any client.

claude mcp add --transport http booking https://mcp.wondel.ai/booking/mcp

One command in your terminal — the server is available in every Claude Code session.

This server works with Claude, Claude Code, Claude Desktop, Claude Cowork, Cursor, VS Code, Windsurf — and any other MCP client. Stdio-only clients connect through npx mcp-remote.

1 tool, fully typed.

Parameters below are transcribed from the server’s own zod schemas — what you see is what the tool validates.

booking_search_hotels Experimental: Booking.com hotels for a location and stay — name, nightly price, review score and count, star class, image and booking link. A call can come back blocked by AWS WAF.
Param Type Description
location * string Free-text location, e.g. "Warsaw" or "Zakopane, Poland".
checkin * date Check-in date, YYYY-MM-DD.
checkout * date Check-out date, YYYY-MM-DD.
adults integer
= 2
Number of adults.
rooms integer
= 1
Number of rooms.
currency string ISO currency code for prices (e.g. PLN, EUR). Omit for Booking’s default for the region.
limit integer
= 25
Max hotels to return (1–50).
no_cache boolean
= false
Force a fresh Booking.com fetch, bypassing this server’s result cache.

Prompts that just work.

Booking.com hotels in Gdańsk’s old town for the last weekend of October, 8.5+ rated, in PLN.

booking_search_hotels

Compare Booking.com’s cheapest well-rated Kraków hotel with what Google Hotels shows for the same dates.

booking_search_hotels

What comes back

booking_search_hotels · illustrative — the shape is the tool’s, the values are examples

{
  "source": "booking.com via firecrawl+brightdata+zenrows",
  "fetchedAt": "2026-09-23T18:30:12.905Z",
  "experimental": true,
  "query": { "location": "Gdańsk", "checkin": "2026-10-30", "checkout": "2026-11-01", "adults": 2, "rooms": 1, "currency": "PLN", "limit": 25 },
  "search_url": "https://www.booking.com/searchresults.html?ss=Gda%C5%84sk&…",
  "currency": "PLN",
  "count": 25,
  "hotels": [
    {
      "name": "…",
      "price_per_night": 412,
      "currency": "PLN",
      "rating": 8.9,
      "reviews": 1284,
      "hotel_class": 4,
      "booking_url": "https://www.booking.com/hotel/pl/…",
      "thumbnail": "https://cf.bstatic.com/…"
    }
  ]
}

Fetching tools share one envelope — source says where the data came from, fetchedAt says when, query echoes what you asked. Compact JSON in one text block, exactly how agents like it.

Add Booking.com
to your agent.

mcp.wondel.ai/booking/mcp

Authless · Read-only public data · Streamable HTTP at /mcp