Chinese-language SERP research
Baiduエンドポイントを使えば、「Chinese-language SERP 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。
Collect Baidu web result cards with titles, destination URLs, snippets, publishers, dates, result types, and pagination, or retrieve Baidu-ranked autocomplete suggestions for Chinese and other prefixes.
Extract Baidu's Chinese search-result cards and ranked suggestions as structured JSON, with page-aware results and typed cards.
エンドポイントファミリー
2
ドキュメント化されたパラメータ
6
サンプル数
2
ライブカタログスナップショット
稼働エンドポイント
2
メソッド
GET
必須パラメータ
4
スキーマ参照
2
{
"platform": "Baidu",
"endpoint": "baidu-search",
"method": "GET",
"path": "/baidu/search",
"auth": "apiKey"
}ユースケース
Extract Baidu's Chinese search-result cards and ranked suggestions as structured JSON, with page-aware results and typed cards.
Baiduエンドポイントを使えば、「Chinese-language SERP 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。
Baiduエンドポイントを使えば、「Publisher and result-type 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。
Baiduエンドポイントを使えば、「Search-query discovery」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Baidu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2件、ドキュメント化されたリクエストパラメータ6個、公開レスポンススキーマ2件）。DocsとPlaygroundが参照しているカタログと同じものです。
Baiduのエンドポイントは2件、2つのリクエストファミリーに分類されています（Search、Suggest）。
これらのBaiduエンドポイントには6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4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。
2件のBaiduエンドポイントのうち2件が実際のサンプルレスポンスを、2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。
Baidu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、429、500、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。
Baiduのエンドポイントは2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
/baidu/search
/baidu/suggest
エンドポイントカタログ
/baidu/searchReturns Baidu result cards for a query: title, destination URL, snippet, publisher, and date, with page-based pagination. Each result's `type` is `web` (standard pages), `video` (video cards), or `baike` (Baidu Baike encyclopedia entries); answer widgets such as weather, entity panels, and recommendation lists are not returned. Baidu challenges a portion of requests with a security verification page; the service retries and returns 503 if every attempt is challenged, so callers should retry on 503.
レスポンスに関する注記
- Each result's `type` is one of `web` (standard pages), `video` (video cards), or `baike` (Baidu Baike encyclopedia entries). Answer widgets such as weather panels, entity panels, Tieba cards without a link, and recommendation lists are not results and are not returned. - `source` and `date` are the publisher name and date Baidu shows on the card; `date` is Baidu's own text, such as `2025年10月30日` or `1月19日`, and is omitted when Baidu shows none. - A single page holds about 10 cards, of which roughly 8 to 10 are returned as results. - Baidu sometimes responds with a security verification page instead of results. If every attempt is challenged the endpoint returns `503`, and the call can simply be retried. - A response page with no result cards and no verification marker is reported as `503` (parser drift), never as an empty success. - `pagination.next_page` is present only when Baidu offers a next page and `page` is below `10`.
MCPツール baidu_search
/baidu/suggestReturns Baidu search-box autocomplete completions for a query prefix, in Baidu's own ranking order. Baidu never returns more than 10 suggestions per prefix; Chinese prefixes are supported.
レスポンスに関する注記
- A prefix with no completions returns a valid empty `suggestions` array, not an error. - Duplicate (case-insensitive) and empty suggestions are removed.
MCPツール baidu_suggest
関連API
Search & Web Discovery
Use a Google Trends API — an API for Google Trends data — to track search demand, trending terms, interest over time, regional interest, related and rising queries, plus Google Suggest discovery, as structured JSON for demand research and trend monitoring.
Search & Web Discovery
Extract Google News results — headlines, source names, article URLs, snippets, thumbnails, and recency — as structured JSON for media monitoring, brand mention tracking, and news aggregation workflows.
Search & Web Discovery
Extract Google Jobs results — job titles, listing snippets, and locations — as structured JSON for job-market intelligence, hiring-trend analysis, and recruitment sourcing workflows.
Baiduのスクレイピング方法
Baidu's company overview describes a mobile ecosystem spanning Baidu App, ERNIE Bot, and Haokan, with AI built alongside its search foundation. Crawlora exposes two narrower public-data contracts: /baidu/search reads normalized cards from Baidu's mobile result page, while /baidu/suggest returns ranked search-box completions. In a live call on 2026-10-07, the query 中国 人工智能 发展 returned eight web cards; Xinhua's 数智跨越 中国人工智能产业发展的“加速度” led page 1 and next_page was 2. Web search uses 1-based pages from 1 through 10; suggestions cap count at 10. Answer widgets are omitted, and verification pages or unexpectedly contentless results are reported as upstream failures, not empty successes.
GET /baidu/search with required q and optional page. Page defaults to 1 and accepts only 1 through 10. Store query and page beside each snapshot so later ranking comparisons stay aligned.
Read each result's position, title, URL, description, source, date, and type. Type distinguishes ordinary web pages, video cards, and Baidu Baike entries; weather panels, entity panels, unlinked Tieba cards, and recommendation lists are intentionally excluded.
GET /baidu/suggest with q and optional count from 1 to 10. The endpoint preserves Baidu's order and removes blank or case-insensitive duplicate suggestions.
Baidu sometimes returns a security-verification page. Crawlora retries across mobile browser profiles and documents 503 when every attempt is challenged; retry with backoff rather than treating that response as a zero-result query.
FAQ
It returns normalized public result cards with position, type, title, URL, snippet, publisher, date when shown, and pagination. Type is web, video, or baike; Baidu answer widgets are outside this endpoint's result set.
Use the 1-based page parameter from 1 through 10. pagination.next_page appears only when Baidu offers another page and the current page is below 10.
Yes. The 2026-10-06 live probe for 人工 returned five ordered Chinese completions, including 人工智能是什么. Count is clamped to a maximum of 10.
The service retries bounded browser profiles and returns a documented upstream failure if every attempt is challenged. Do not record that as an empty SERP. The bounded 2026-10-06 live search probe exceeded 12 seconds, while the suggestion endpoint succeeded.