WeChat Search

API docs
https://api.sandbase.ai/v1/api/wechat-search/v2/search

Search WeChat with one endpoint: official account articles, accounts, Channels videos, live streams, news and more. Filter by vertical, sort and publish time, with cursor pagination.

Price
$0.01USD / call
Method
POST
Response
JSON
Parameters
1 required · 6 optional
100 free calls / dayFirst 100 calls per day free (shared per organization) · Resets at 00:00 UTC. Pay per call after that.

Playground

Input
Search keyword (1-100 chars)
Optional, default `all`. Vertical string key, only the following are supported — `all` / `account` / `article` / `video` / `live_stream` / `moments` / `news` / `book` / `listen` / `image` / `encyclopedia` / `weixin_index`; any other value is rejected (422).
Optional. Pagination cursor. Leave empty for the first page; for the next page pass back the `cursor` returned in the previous response (repeat the same `sort` / `publish_time`), and use `continue_flag` to check whether there are more pages.
Optional, default 0 (>=0). Pass `0` for the first page; **use `cursor` to paginate — offset alone does not work** (it returns the first page every time). Minimum: 0.
Optional, default `all`. Publish time (result page "time" dropdown, common) — `all`/0 / `day`/1 / `week`/2 / `half_year`/3; string key or integer, invalid values rejected (400).
Optional, default True. True=raw search response; False=simplified parsing.
Optional, default `default`. Sort (result page "sort" dropdown, common to all verticals) — `default`/0 (relevance) / `latest`/1 (newest) / `hot`/2 (most liked); string key or integer, invalid values rejected (400).
OutputExample
Responseoutputs · 3.9 KB · 88 lines
[
  {
    "data": {
      "business_type": 2,
      "categories": null,
      "continue_flag": 1,
      "count": 37,
      "cursor": "uIPAVvHcM7SXlk126cnQ4rzNSw_0qdP-8OVyX2TzuPWKBg1BLPozHpQvhQYD6dWC5hoFSF8A95LzJ4crMiVoupQk-eQm6Bq2yD23YCmHM315GM1atXV__HdN7rft2SSilzYt9aryqD_UegKjKhODC0rv_Oo4PYhnEZ4mYK77_uk3LIHEjRZeA-iz-Vot3cBn6F55Ib8QURgJSJP4t2hocZ0eBPa-LHwyYuhxYeQzA4SDYHkl_luqoiuX5Q6q-moqxVcAAzKIz1qyLwkU2fHAOvgcy_xhVaZiHCdsgI9awOGHO_DQdOsYazvaha5TZhfZqkho9rPF18maOsn3x0ZaKOHZC_MVTPhn3SE3ViEuq-X5VFVNqmQF-8N05tYjSeIbTafatpNuChYQI7Cp5qYNF4rpFt_5hUBzYVWdw1Fj0p-OvJN4S-M1S-xWWJVzQc8EDtYGLiuVWbQwlKAdXHsDiNBabplV7UomiiJNEzq_p5ZbVENbtfNDhrA4OnOC33rBcvSf_XWR4IilAtQYuQdxEuY2n2juOzDsoEcb6uVDb05tbEfvVsAZobkOl2Cib292vq35mtb8NyeuAHv92fdR4658…",
      "debug_id": "e295d4ab",
      "debug_info": "gAAAAABqxjABtsna3kB6S0dvg_DtrLriNL9xJqHje5kvhN9X2EpTauuMK-Met11Sv3mIBTYThJOXsgwmdYbSY_FIgp1JaBPMxA7zhOu4te4jzA9NOhX_zXKJIAacFG_qCtdMlPz4R0w5n4XebHyuuPl6fVcj9gDnd2ENHRH496wsbjYaZkGt0UM_0FZKCgo-sfYHMWzcCJ7yH4BgiFUgQdGtoVk36QUeqk9Eoa9yieuYe6IP_DhwVbZfnppcDfWUE4OYs3msr7nR",
      "items": [
        {
          "authStatus": [
            {
              "desc": "「搜索服务」无定位信息,默认展示位置为北京的结果",
              "title": "根据定位可提供更精准的结果",
              "type": 1
            },
            {
              "pop": {
                "desc": "若需获得所在地相关结果,可",
                "descAction": {
                  "jumpInfo": {
                    "api": "requestSystemCapabilityAuthorization",
                    "apiParams": "{\"capability\":1,\"authScene\":11}",
                    "jumpType": 14
                  },
                  "title": "打开定位服务"
                },
                "overSeaTitle": "「搜索服务」无当前定位信息, 已展示北京相关结果",
                "title": "「搜索服务」无位置权限, 已展示北京相关结果"
              },
              "title": "无当前定位信息,默认展示北京相关结果",
              "type": 2
            },
            {
              "overSeaTitle": "根据当前定位可提供更精准的结果",
              "title": "根据定位可提供更精准的结果",
              "type": 3
            }
          ],
          "docID": "addrsupportTip",
          "showType": 1,
          "title": "允许使用【位置】,获得更精准的搜索结果"
        },
        {
          "date": 1766199600,
          "desc": "",
          "docID": "6336858824649249438",
          "docType": 0,
          "doc_url": "http://mp.weixin.qq.com/s?__biz=MzI2ODUzNzA4NQ==&mid=2247581064&idx=1&sn=c6ddd50b6a647a3210cdb2b4eafe7552&chksm=eba93e1c4690a9f446de1affca2e5aebbebc9ea682da1e4b6ee26f1a912bf302db5b33560b0d#rd",
          "itemShowType": 0,
          "mpScene": 7,
          "source": {
            "dateTime": "10个月前",
            "title": "青苗法鸣"
          },
          "src_type": 49,
          "thumbUrl": "https://mmbiz.qpic.cn/mmbiz_jpg/bewibHHVTya7wjLl1P83s4jXluaUib28XNDJjmbcu9ezvCoaYiakpsw4L9b4Of9PFWTFv2f4tL9dARHAIs0yQZNTw/0?wx_fmt=jpeg",
          "timestamp": 1766199600,
          "title": "<em class=\"highlight\">2026</em>年<em class=\"highlight\">南开大学法学</em>/法律<em class=\"highlight\">博士拟录取名单</em>"
        },
        {
          "date": 1784633400,
          "desc": "",
          "docID": "688829883682389426",
          "docType": 0,
          "doc_url": "http://mp.weixin.qq.com/s?__biz=MzYzMTI4MDcyOA==&mid=2247490020&idx=4&sn=4db353f9909dfaee928b3de05b1afb7e&chksm=f1e42329bd9ab957ba86ff25a6780aed9fd15082599b71e6a2daa0600f181abd133460a50915#rd",
          "itemShowType": 0,
          "mpScene": 7,
          "source": {
            "dateTime": "3个月前",
            "title": "经管法申博圈-青于蓝考博"
          },
          "src_type": 49,
          "thumbUrl": "https://mmbiz.qpic.cn/mmbiz_jpg/ErbuWleBsMl4FObWeT7NsYFQOfkVsPDG3Vvk2iaRfKxdavASAKEGNQkVNG7LtYVDepSEoLEJsB6JvCRMZlmIqCLpicL7xPwzSGj5Oc46gsYbY/0?wx_fmt=jpeg",
          "timestamp": 1784633400,
          "title": "<em class=\"highlight\">南开大学法学院2026</em>年<em class=\"highlight\">博士生拟录名单公示</em>"
        },
        "… 34 more items in the full response"
      ],
      "keyword": "南开大学 法学院 2026 博士 拟录取名单",
      "no_more": null,
      "offset": 36,
      "total": 243
    }
  }
]

Examples

Select an example to load its inputs and result into the playground.

API README

WeChat Search API

Searches WeChat for a keyword and returns matching results from the vertical you choose with business_type, such as official account articles, accounts, Channels videos, live streams or news. It needs one input, keyword. Filters for sort order and publish time narrow the results.

Use cases

  • Track new official account articles about a brand or topic with sort set to latest.
  • Collect article titles, snippets and links for a keyword to build a media monitoring feed.
  • Review publisher display names within the returned sample; this does not establish which accounts publish most on the subject.
  • Limit a search to the last day or week with publish_time to check recent coverage.

Request

curl -X POST "https://api.sandbase.ai/v1/api/wechat-search/v2/search" \
  -H "Authorization: Bearer $SANDBASE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"keyword": "OpenAI", "business_type": "article", "raw": false}'

The result is returned in outputs[0].data.

Parameters

NameTypeRequiredDescription
keywordstringYesSearch keyword (1-100 chars)
business_typestringNoOptional, default all. Vertical string key, only the following are supported — all / account / article / video / live_stream / moments / news / book / listen / image / encyclopedia / `weixin_index
cursorstringNoOptional. Pagination cursor. Leave empty for the first page; for the next page pass back the cursor returned in the previous response (repeat the same sort / publish_time), and use continue_flag to check whether
offsetintegerNoOptional, default 0 (>=0). Pass 0 for the first page; use cursor to paginate — offset alone does not work (it returns the first page every time).
publish_timestringNoOptional, default all. Publish time (result page "time" dropdown, common) — all/0 / day/1 / week/2 / half_year/3; string key or integer, invalid values rejected (400).
rawbooleanNoOptional, default True. True=raw search response; False=simplified parsing.
sortstringNoOptional, default default. Sort (result page "sort" dropdown, common to all verticals) — default/0 (relevance) / latest/1 (newest) / hot/2 (most liked); string key or integer, invalid values rejected (400).

Example response

Real response for the request above, trimmed: lists keep their first item and long strings are cut.

{
  "business_type": 2,
  "categories": null,
  "continue_flag": 1,
  "count": 38,
  "cursor": "hO0SuC6V7gEWmcGXQvsJ5wYWU4HyplzRxHlbePoYMtHLQY0vwElfi8d9LA-X87FzUQ7qOvpJTpv4SPHP…",
  "debug_id": "e5835ac5",
  "debug_info": "gAAAAABq6ZE7ot2LojpCJR304JESp7F_lWS4y7oOLUfqwVqQ9TIO8WNyZlsFYGWjOnT9jqzEk27z2sgB…",
  "items": [
    {
      "date": 1793688900,
      "desc": "<em class=\"highlight\">OpenAI</em>不是一家普通的公司。GPT-6 Astra、5亿年化营收、全球最大的AI用户入口,这些东西是真…",
      "docID": "5774067735576686510",
      "docType": 0,
      "doc_url": "http://mp.weixin.qq.com/s?__biz=MzU4ODQwNTIxMw==&mid=2247595037&idx=1&sn=8c2c2d5…",
      "itemShowType": 0,
      "mpScene": 7,
      "source": {
        "dateTime": "51分钟前",
        "title": "人工智能产业链union"
      },
      "src_type": 49,
      "thumbUrl": "https://mmbiz.qpic.cn/sz_mmbiz_jpg/9NdMVNE3GyZlQFyh70as5OnDBo1sPia1hpUW8PMhpoHnf…",
      "timestamp": 1793688900,
      "title": "【投融资】没人抢了:<em class=\"highlight\">OpenAI</em>的估值,一周之内自己砍掉了1500亿美元"
    }
  ],
  "keyword": "OpenAI",
  "no_more": null,
  "offset": 38,
  "total": 280
}

Response fields

FieldMeaning
items[].titleArticle title. Search matches are wrapped in <em class="highlight"> tags.
items[].descShort text snippet of the article, with matched keywords wrapped in <em class="highlight"> tags.
items[].doc_urlLink to the article on mp.weixin.qq.com.
items[].source.titleName of the official account that published the article.
items[].source.dateTimeRelative publish time as shown on the WeChat result page, for example 51分钟前 or 1天前.
items[].timestamp / items[].datePublish time of the article as a Unix timestamp in seconds.
items[].thumbUrlCover image URL for the article.
items[].docIDIdentifier of the result document.
count / totalcount is the number of items in this page's result. total is the overall result count reported by the search.
continue_flag / cursorcontinue_flag is 1 when more pages exist. cursor is the value to pass back to get the next page.

Notes

  • To get the next page, pass the returned cursor back with the same keyword, sort and publish_time. Check continue_flag first.
  • offset alone does not paginate. It returns the first page every time, so use cursor.
  • sort accepts default, latest or hot. publish_time accepts all, day, week or half_year.
  • For this endpoint, select business_type account, article or video explicitly and inspect that vertical’s response before parsing it.

Pricing

$0.01 per successful call. Failed calls are not charged. It is included in the daily free quota of each organization.

Search one page of Official Account articles

Send your keyword with business_type article and raw false explicitly. The default vertical is all and the default raw value is true, so omitting these fields changes the task and response shape. Keyword length is 1–100 characters. Start with offset 0 and no cursor. Use the documented sort and publish_time choices; latest sorts by recency, while default ranks relevance and hot is the provider's popularity ordering.

Read items as a list. A present empty list means no items were returned by this request; a missing list is not the same result. Each item can contain title, desc, doc_url and source.title. The title and snippet can include highlight markup. Extract plain text or sanitize HTML before display; a highlighted snippet is not the full article body. A source display name alone is not a stable account identity.

Preserve the returned publication timestamp when available, its raw value and retrieval time. Relative source.dateTime labels such as minutes ago cannot establish an exact publication time without observation context. Search count and total are provider-reported values, not proof you obtained every matching article or all articles from an account. The README excerpt keeps one item while showing count 38, so its displayed length is not a live result count.

A later page needs the returned cursor and a continuation decision from continue_flag, while keeping keyword, business_type, sort, publish_time and raw unchanged. Offset alone repeats the first page; do not implement an offset loop or reuse the truncated example cursor. No continuation roundtrip was tested for this revision. For accounts or Channels content, select another supported business_type and inspect that response separately rather than assuming article fields apply.

Choose a complete mp.weixin.qq.com article URL before a separate Article Detail request. Never send the ellipsized URL from the README example, automatically open every hit, or treat search titles as verified claims about the subject.

Give your agent one article task

Use a local SANDBASE_API_KEY with Bearer authentication and POST /v1/api/. An agent that already has access needs your chosen keyword or complete article URL and one desired output. For a new project, share the setup Skill, Markdown quickstart and selected endpoint. Read one result before expanding to more searches, article bodies or scheduled monitoring.

Validate completed SandBase status and an object at outputs[0].data. The simplified reference places search items or the article content object directly there; do not add a guessed data.data wrapper or an undocumented provider success flag. A missing or malformed required object/list is an error, not an empty successful research result. The shortened reference JSON is evidence of selected fields, not a complete field guarantee.

Pricing checked on 2026-11-08: both endpoints list $0.01 per call and eligibility for 100 daily calls shared across eligible data APIs in the organization, resetting at 00:00 UTC. This is not a remaining-balance or allowance measurement. Verify current pricing before running. No new business request was made for this revision. Follow the WeChat Python integration guide for bounded parsing and dated source notes.

Explore more

Related APIs

All APIs
APIPriceResponse
WeChat Search$0.01 per callJSON
WeChat Search$0.01 per callJSON
WeChat Search$0.114286 per callJSON
WeChat Search$0.114286 per callJSON
WeChat Search$0.057143 per callJSON

Questions

FAQ

How do I use the WeChat Search API?

Send a POST request to https://api.sandbase.ai/v1/api/wechat-search/v2/search with your SandBase API key in the Authorization header (Bearer) and the parameters as a JSON body. You can also try it in the Playground on this page without writing code.

How much does the WeChat Search API cost?

$0.01 per call. Every organization gets 100 free calls per UTC day, shared across eligible SandBase APIs including this one; after that each call is billed at this price from your SandBase balance. There is no subscription.

What parameters does the WeChat Search API accept?

Required: keyword (string). Optional: business_type (one of all, account, article, video, live_stream, moments, news, book, listen, image, encyclopedia, weixin_index), cursor (any), offset (integer), publish_time (any), raw (boolean) and sort (any). The Parameters table and the Playground on this page list every field.

What does the WeChat Search API return?

A JSON response returned in the same HTTP request (synchronous). The README on this page includes a sample response.

Do I need a WeChat Search developer account or API key?

No. WeChat Search runs through SandBase, so one SandBase API key covers this endpoint and every other API and model in the catalog. You do not need to apply for WeChat Search API access separately.

Which options select simplified article search?

Send keyword, business_type article and raw false explicitly. The defaults are all and raw true. Start with offset 0 and no cursor; other verticals and raw responses can have different shapes.

Can I get the next page by increasing offset?

No. The schema states that offset alone returns the first page again. A later request uses the returned cursor and continue_flag with the same query, vertical and filters. The truncated reference cursor is unusable, and this revision did not test a continuation roundtrip.

Does a search hit contain the full article and exact account identity?

The simplified article result describes titles, snippets and links, sometimes with HTML highlight tags. It is not the full article body, and a publisher display name is not stable account identity. Select a complete article URL for a separate detail request.