Restaurant discovery and menu monitoring
Zomatoエンドポイントを使えば、「Restaurant discovery and menu monitoring」をドキュメント化された入力とJSONレスポンスによる再現可能なAPIリクエストに変えられます。
Turn Zomato's public restaurant data into structured JSON — a city-wide listing with popularity, rating and cost sorting, pure-veg filtering and cursor pagination, keyword search within a city, Zomato's own curated editorial collections such as Best pubs & bars or Iconic restaurants, one collection's restaurant list, one restaurant's public detail by canonical URL, and its menu grouped into categories. Credential-free.
Search Zomato restaurants by city with sort and cursor paging, browse curated collections, and get restaurant detail and menus as JSON.
エンドポイントファミリー
4
ドキュメント化されたパラメータ
14
サンプル数
5
ライブカタログスナップショット
稼働エンドポイント
5
メソッド
GET
必須パラメータ
10
スキーマ参照
5
{
"platform": "Zomato",
"endpoint": "zomato-search",
"method": "GET",
"path": "/zomato/search",
"auth": "apiKey"
}ユースケース
Search Zomato restaurants by city with sort and cursor paging, browse curated collections, and get restaurant detail and menus as JSON.
Zomatoエンドポイントを使えば、「Restaurant discovery and menu monitoring」をドキュメント化された入力とJSONレスポンスによる再現可能なAPIリクエストに変えられます。
Zomatoエンドポイントを使えば、「India dining market and cuisine research」をドキュメント化された入力とJSONレスポンスによる再現可能なAPIリクエストに変えられます。
Zomatoエンドポイントを使えば、「Local hospitality competitive analysis」をドキュメント化された入力とJSONレスポンスによる再現可能なAPIリクエストに変えられます。
マネージド実行
以下の数値はすべて、稼働中のZomatoエンドポイントカタログから取得しています（エンドポイント5件、ドキュメント化されたリクエストパラメータ14個、公開レスポンススキーマ5件）。DocsとPlaygroundが参照しているカタログと同じものです。
Zomatoのエンドポイントは5件、4つのリクエストファミリーに分類されています（Restaurant、Collection、Collectionsほか1件）。
これらのZomatoエンドポイントには14個のリクエストパラメータがドキュメント化されており、うち10個が必須です。統合コードを書く前に入力仕様をすべて確認できます。
5件のZomatoエンドポイントのうち5件が実際のサンプルレスポンスを、5件がドキュメント化されたレスポンススキーマを備えています。最初のリクエストの前に実データの形に合わせて実装できます。
Zomatoのエンドポイントは成功時のスキーマに加えてエラーレスポンス（400、404、503）もドキュメント化しています。ブロック、レート制限、レコード欠損は空データではなく型付きのエラーとして返ります。
Zomatoのエンドポイントは5個のホスト型MCPツールとして提供されており、エージェントは同じパラメータと同じJSON契約のまま、追加の実装なしに同じルートを呼び出せます。
カバレッジマップ
これらのカードは稼働中のエンドポイントカタログから生成されるため、このページはDocsとPlaygroundが使うAPIサーフェスをそのまま反映します。
/zomato/restaurant
/zomato/collection
/zomato/collections
/zomato/search
エンドポイントカタログ
/zomato/searchReturns restaurants in a Zomato city, optionally filtered by a keyword. With no query this reads Zomato's own city-wide delivery listing, which pages through an opaque cursor (~12 restaurants per page: pass a response's next_cursor back as cursor until has_more is false, which marks the end of the city's listing) and accepts Zomato's own sort and Pure Veg filters; with a query it uses Zomato's own keyword search across restaurant names, dishes, and cuisines, which returns a single unpaginated result set. Each restaurant carries its url (the value the restaurant and menu endpoints take), name, cuisines, cost for two, rating, distance from the city location, delivery time, whether Zomato currently delivers from it, whether the placement is promoted (sponsored), and — on the plain city-wide listing — Zomato's own headline offer badges.
MCPツール zomato_search
/zomato/restaurantReturns one Zomato restaurant's public detail by its canonical URL: name, address, locality, city, postcode, coordinates, phone, cuisines, headline rating plus Zomato's separate dining and delivery rating aggregates, the full weekly opening-hours table, open/closed and delivery-only/dark-kitchen status, and delivery metadata (ETA, minimum order, pickup availability). Zomato has no lighter restaurant-detail-only source -- this reads the same upstream as /zomato/restaurant/menu, just without the menu items.
MCPツール zomato_restaurant
/zomato/restaurant/menuReturns one restaurant's menu grouped into categories, plus its restaurant summary. A category is named by Zomato's own sub-heading when it has one (e.g. "Veg Burgers") and otherwise by the menu section it sits under (e.g. "Burgers"), since most restaurants name only the section level. Each item carries a name, description, veg/non-veg flag, image, and price when Zomato's own anonymous response includes one -- confirmed live, item prices are commonly gated behind login for the delivery-ordering flow, so price is frequently absent even though the item itself is public.
MCPツール zomato_restaurant_menu
/zomato/collectionsReturns Zomato's own curated "best of" restaurant collections for a city (e.g. "Best pubs & bars", "Iconic restaurants", "Insta-worthy spots") -- distinct editorial groupings of restaurants, not a keyword/geo search. Each collection's url is the value /zomato/collection takes. Zomato curates collections for larger cities only; a real city it curates none for returns a 404.
MCPツール zomato_collections
/zomato/collectionReturns one Zomato curated collection's restaurant list (name, cuisines, locality, rating, and headline review count where Zomato's own response includes them). Reads the same underlying page-render source as /zomato/restaurant and /zomato/restaurant/menu, pointed at the collection's own page instead of a restaurant's.
MCPツール zomato_collection
関連API
Marketplaces & Retail
Collect marketplace product signals from Amazon without building brittle storefront scrapers.
Marketplaces & Retail
Build resale, pricing, and marketplace workflows from structured eBay data.
Marketplaces & Retail
Turn public Shop.app product and merchant pages into structured JSON for e-commerce product intelligence, price research, shop monitoring, and marketplace discovery workflows.
Zomatoのスクレイピング方法
Crawlora's Zomato endpoints return normalized public Zomato data as JSON with one API key — no Zomato account required. Zomato's city-wide listing and its keyword search are two different upstream surfaces, and only the listing supports sorting, filtering, and pagination.
The search endpoint takes a Zomato city slug — the first path segment of a zomato.com city URL, such as mumbai or delhi-ncr. With no keyword it reads Zomato's own city-wide directory, which supports sorting by popularity, rating or cost, a pure-veg restriction, and cursor pagination at roughly twelve restaurants per page. Add a keyword to search restaurant names, dishes or cuisines instead — that path returns a single unpaginated, unfiltered result set, because Zomato's own keyword surface works that way.
The collections endpoint returns Zomato's own editorial groupings for a city — Best pubs & bars, Iconic restaurants, Insta-worthy spots and the like. These are distinct curated lists Zomato surfaces on the city's collections hub, not a keyword or geo search, and each collection's URL is what the collection endpoint takes to return that list's restaurants.
The restaurant endpoint returns a single restaurant's public detail by its canonical Zomato URL, which every listing, search result and collection entry carries.
The restaurant-menu endpoint returns one restaurant's menu grouped into its own categories, alongside the restaurant summary — the same page-render source the detail and collection endpoints read, pointed at the menu instead.
FAQ
Call the search endpoint with a Zomato city slug to list or keyword-search that city's restaurants, browse the curated collections for editorial groupings, then pass a restaurant's canonical URL to the restaurant and restaurant-menu endpoints for its detail and menu — all return structured JSON with one API key.
Through the city-wide listing, not keyword search. Leave the keyword off and the listing returns roughly twelve restaurants per page with a has_more flag and a next_cursor to pass back as cursor. Sorting and the pure-veg filter also apply only on this path — Zomato's keyword-search surface returns a single unpaginated, unfiltered set.
The first path segment of a zomato.com city URL — mumbai, delhi-ncr and so on. It is required on the search endpoint, and an unrecognized slug returns a 404 rather than an empty result. A recognized city that simply has no matches returns an empty restaurant list, which is a normal success.
Yes — the collections endpoint returns the editorial groupings Zomato itself publishes on a city's collections hub, such as Best pubs & bars or Iconic restaurants. Each carries the URL that the collection endpoint takes to return that collection's own restaurant list, which reads the same underlying page-render source as restaurant detail.
Yes — the restaurant-menu endpoint returns one restaurant's menu grouped into its own categories together with the restaurant summary. Take the restaurant's canonical URL from any listing, search result, or collection entry and pass it there; the same URL is what the restaurant detail endpoint takes.