WeChat Official Account Scraper 公众号 - Articles

The WeChat Official Account Scraper collects public 公众号 articles (微信公众号, Weixin) with no login and no WeChat account. Search Sogou's WeChat index by keyword, read any mp.weixin.qq.com article link in full, or export every article in an account's album.

中文说明

微信公众号（Weixin）文章抓取工具，数据来自搜狗微信搜索与 mp.weixin.qq.com 文章页。五种模式：按关键词搜索公众号文章、按链接获取文章全文、获取公众号合集（专辑）中的全部文章、按关键词查找公众号（account_search）、按微信号查看搜狗已收录的文章（account_articles）。

搜索结果含标题 title、摘要 summary、公众号名称 account_name、发布时间 published_at 与永久链接；开启全文后每行还含正文 content_text、Markdown、图片、作者、原创标记与合集信息。

无需登录，也无需微信账号。运行自带住宅代理，结果支持导出 JSON、CSV 或 Excel。阅读数、点赞数与评论不对匿名访问开放，因此不提供。

account_search 每个公众号返回一行：名称、微信号 wechat_id、是否认证、命中篇数与最新一篇的标题和时间（不含链接）。部分公众号没有自选微信号，wechat_id 显示为 gh_ 开头的原始 ID，这类账号不能用于 account_articles。account_articles 只返回搜狗已收录的部分文章，是样本而非完整历史，可能不全、无序、偶尔为空，每个微信号最多约 95 行，只对匹配的行计费。

What you get

One dataset, one row per record, with kind saying what each row is. A search result row:

{
  "kind": "search_result",
  "source_keyword": "咖啡",
  "page": 1,
  "position": 1,
  "total_results_estimate": 9455,
  "docid": "ab735a258a90e8e1-6bee54fcbd896b2a-38aaebf7e189d78e56f37bdd146f4cf5",
  "title": "咖啡地图 | “咖啡”遇上西雅图,好喝夜未眠",
  "summary": "不久前, 君去西雅图出差,参加咖啡展会.短短几天内,一口气逛了40多家咖啡馆……",
  "account_name": "企鹅吃喝指南",
  "account_username": "qiechihe",
  "account_verified": true,
  "published_at": "2017-06-11T11:52:21+08:00",
  "msgid": "2653009417",
  "idx": 1,
  "permanent_url": "https://mp.weixin.qq.com/s?__biz=MjM5Mzc5NTk1OQ==&mid=2653009417&idx=1&sn=147bc1880d7a81c3e16e91cb88f10377&chksm=bd448c11…",
  "cover_url": "http://mmbiz.qpic.cn/mmbiz_jpg/ibLgUdQXiaIVQkJlRvYZaRXj…/0?wx_fmt=jpeg",
  "scraped_at": "2026-09-27T17:11:39+00:00",
  "_meta": {"author": "https://apify.com/herus13", "status": "ok", "message": null}
}

_meta is run metadata, not scraped data. A target that fails appears as a row with only _meta (status failed, the reason in message); those rows are never charged.

Every mode writes into the same dataset, and the Output tab has one view per kind:

• account — one Official Account found by account_search: account_name, wechat_id, account_openid, verified, avatar_url, articles_found (how many of this search's hits came from the account), latest_article (its title, published_at and published_timestamp, with no link) and source_keyword.
• search_result — one search hit: title, summary, account_name, account_username (the account's WeChat id), account_verified, published_at, permanent_url, cover_url, has_video, the hit's page and position, and source_keyword.
• album_article — one article of an album: title, url (its permanent link), published_at, position in the album, cover_url, plus the album's album_title, album_article_count and album_read_count.
• article — one full article: title, author, digest, content_text, content_markdown, content_html, images (with width and height), video_ids, published_at, the account (account_name, account_username, account_alias, account_signature), is_original, the region the account chose to show (ip_region_province), its album and tags, and read_more_url (阅读原文). A full article found by search carries the search hit in search; one found in an album carries the entry in album_entry.

An account row:

{
  "scraped_at": "2026-10-06T19:41:51+00:00",
  "kind": "account",
  "source_keyword": "跨境电商",
  "account_name": "亿恩",
  "wechat_id": "enecnews",
  "account_openid": "oIWsFt0xHoC1sCSZbq-E4UiC-6qw",
  "verified": true,
  "avatar_url": "http://wx.qlogo.cn/mmhead/Q3auHgzwzM5P6vwhy46wzNUAHH5bSflGtWgKK60UZcyB5mibg3hiattA/0",
  "articles_found": 1,
  "latest_article": {
    "title": "部分卖家卖房还债!跨境电商“坑害”一批普通人",
    "published_at": "2026-10-06T11:32:58+08:00",
    "published_timestamp": 1791257578
  },
  "_meta": {
    "author": "https://apify.com/herus13",
    "status": "ok",
    "message": null
  }
}

account_search returns one row per distinct account among a keyword's search hits, reading up to ten search pages before it applies max_items. An account found by two keywords gets one row for each. articles_found counts hits in this search, not the account's total articles. account_articles returns the same article rows as search, plus account_id, only for hits whose account ID equals the WeChat ID you gave. It is what search has indexed for that ID: a partial, unordered, possibly stale sample of about 95 rows at most, not the full history, and an ID that search has not indexed returns no rows and a notice.

Times are ISO 8601 in Beijing time. permanent_url is set whenever the source gives a permanent link. On search rows that is only some hits: Sogou gives most results as an encrypted sogou_url that expires within hours, with no permanent link behind it until the article is opened, so read those promptly or turn on include_content. A full article carries biz, mid and idx whenever the page or its link names them; sn and permanent_url need the article's permanent link, which a search hit without one does not reveal, so they stay empty there.

What it costs

Event	Price
Actor Start (apify-actor-start)	$0.00005
search result (search-result)	$0.003
album article (album-article)	$0.003
article (article)	$0.015

Input

Field	Type	Required	Default	What it does
mode	string	no	"search"	What to collect: search finds Official Account articles by keyword; article reads specific article links in full; album lists every article in an account's album (合集); account_search lists the distinct accounts behind a keyword's search results; account_articles lists the articles WeChat search has indexed for the WeChat IDs you give. Each mode reads only its own fields — a field filled in for another mode is rejected, not ignored, so clear it when you switch. Default search.
keywords	array	no	—	Search terms, one per entry, for example 人工智能 or 咖啡. Each term returns up to max_items articles; an article found by two terms is returned once. Required when mode is search or account_search; used by search and account_search — leave empty for every other mode (the form's example term must be cleared).
article_urls	array	no	—	mp.weixin.qq.com article links, one per entry: a short link (https://mp.weixin.qq.com/s/ plus an id), a full link carrying __biz, mid, idx, sn and chksm, or a search-result link (/s?src=11&timestamp=…&signature=…). A full link without chksm is refused, since WeChat answers it with a verification page instead of the article. Required when mode is article; used ONLY by article — leave empty for every other mode.
album_urls	array	no	—	WeChat album (合集) links, one per entry, such as https://mp.weixin.qq.com/mp/appmsgalbum?__biz=…&album_id=…. Each album returns up to max_items articles, newest first, with the permanent link of each. Required when mode is album; used ONLY by album — leave empty for every other mode.
wechat_ids	array	no	—	WeChat IDs as an account's profile shows them, one per entry, for example rmrbwx or huxiu_com. Not the gh_ original ID and not the display name: both are refused, never searched. Only articles whose account ID matches exactly are returned: what WeChat search has indexed for the ID, a partial, unordered, possibly stale sample, not full history. Required when mode is account_articles; used ONLY by account_articles — leave empty for every other mode.
max_items	integer	no	20	How many rows to return per keyword, per album or per WeChat ID (account rows for account_search). N returns at most N; 0 returns none; -1 means no limit. Search stops at WeChat search's own depth, about 94 articles per keyword, whatever the limit; an album is read up to 2,000 articles. Article mode must keep the default. Default 20.
include_content	boolean	no	false	Also open every search hit or album entry and return the full article instead of the summary row: body text, HTML and Markdown, images, author, account details, originality and tags. One more request per row, billed as an article. Default false. Used by search and album; article mode always returns full content and must leave it off.
since	string	no	—	Keep only articles published on or after this date, written as YYYY-MM-DD in Beijing time, for example 2026-09-01. Leave empty for no lower bound. WeChat search cannot filter by date and orders by relevance, so the filter runs on the rows fetched: a narrow window can return fewer rows than max_items. Used by search, album, account_search and account_articles — leave empty for article mode.
proxyUrls	array	no	—	Leave empty and the run uses the residential proxy this actor ships with, included in the price of the run. To route the run through your own account instead, add one gateway URL per entry, for example http://user:pass@host:port — works with any provider that issues URLs. When set, the run uses these first and switches to the included proxy if they stop working, so a broken proxy never fails the run. Use US exits.

How to run it

1. Open the actor and pick a What to scrape mode. Each mode reads only its own fields; clear the example keyword if you switch to article or album.
2. Fill that mode's target: keywords for search, article links for article, album links for album.
3. Turn on Include full article content if you want the body of every search hit or album entry, not just the summary row.
4. Run it and export the dataset as JSON, CSV or Excel.

Full articles for two keywords published since 1 September:

{
  "mode": "search",
  "keywords": ["人工智能", "新能源汽车"],
  "max_items": 30,
  "include_content": true,
  "since": "2026-09-01"
}

Every article in an album:

{
  "mode": "album",
  "album_urls": ["https://mp.weixin.qq.com/mp/appmsgalbum?__biz=MjM5MjA0MDk2MA==&action=getalbum&album_id=3144682336811220998"],
  "max_items": -1
}

Minimal input:

{
  "mode": "search",
  "keywords": ["咖啡"],
  "max_items": 5
}

Available as the MCP tool herus13--wechat-scraper on mcp.apify.com and through the Apify API; send the same JSON.

Use cases

• Brand and product monitoring: search a brand or product name every day and collect which Official Accounts wrote about it, with titles, summaries and publish times.
• Content research and archiving: keep the full text, Markdown and images of an account's articles by reading its albums, for search indexes, knowledge bases or LLM pipelines.
• Media and KOL discovery: find which accounts publish most on a topic and whether they are verified, then read their albums for their back catalogue.
• Market and policy research on China: gather what WeChat's publishers, from state media to industry newsletters, say about a sector, as plain text ready for translation or a classifier.
• Finding accounts to follow or monitor: search a topic with account_search to list the distinct Official Accounts writing about it, with their WeChat ID, verified flag and latest article, then watch the ones that matter.
• Link enrichment: turn a list of mp.weixin.qq.com links shared in chats or spreadsheets into titled, dated, full-text rows.

FAQ

Do I need a WeChat account? No. Every mode reads what an anonymous visitor sees, and the actor never logs in.

How many results does a keyword return? WeChat search shows an anonymous visitor ten pages per keyword, about 94 articles, and ranks them by relevance. Use several related keywords to go wider, and albums to go deeper into one account.

Can I get read counts, likes or comments? No. WeChat serves read, like and 在看 counts and comments only inside its own app to a signed-in user, so no anonymous scraper can return them honestly. They are left out rather than shipped as zeros.

Can I get an account's full history? Not directly: WeChat keeps an account's history page inside its app. Albums (合集) are public, so an account's albums list its articles back to the start, with permanent links.

Why was my article link refused? A full article link needs its chksm part; without it WeChat answers with a verification page instead of the article. Copy the whole link from the browser's address bar or use the short mp.weixin.qq.com/s/… link.

Why did my run stop with "does not use"? Each mode reads only its own fields, and a field filled in for another mode is refused rather than silently ignored. The form opens with an example keyword; clear it when you pick article or album.

Why does since return fewer rows than max_items? WeChat search cannot filter by date for an anonymous visitor, so the date filter runs on the results fetched. A narrow window keeps only the matching ones.

What is account_search for? It lists the distinct Official Accounts behind a keyword's search results, one row per account, so you can find who writes about a topic: name, WeChat ID, verified flag, how many of the hits came from the account and its latest article. It reads up to ten search pages per keyword before applying max_items.

Why does account_articles return a partial sample, or nothing? It returns the articles WeChat search has indexed for the WeChat ID, filtered to rows whose account ID matches exactly. That is a partial, unordered, possibly stale sample of about 95 rows at most, not the full history. In a probe of six IDs, four returned mostly their own articles (68 to 95 rows), one only 4 of 92 hits, and one none. An ID that search has not indexed gives no rows and a notice. For a full back catalogue use albums.

Why do some accounts show a gh_ ID, and why can't account_articles take it? Some of the accounts account_search finds (34 of 87 for 人工智能, none of 49 for 跨境电商) have no chosen WeChat ID, so their wechat_id is the gh_ original ID. account_articles needs the ID the profile shows, such as rmrbwx, and refuses a gh_ ID or a display name with a named error rather than searching.

What does an account row cost? An account row is billed as a search result, $0.003 per row. account_articles is billed $0.003 only for each matching article row; rows that do not match the ID are not charged.

Is a proxy included? Yes, a residential proxy is included in the price of the run. To use your own, paste gateway URLs into Your own proxy URLs.

What happens if my input is wrong? The run stops before it scrapes anything, so no results are charged. Its status and log name each field that is wrong and say what to change — fix the input and run again.

Is scraping WeChat legal? This actor reads only public articles, the ones any visitor can open in a browser, and never signs in. You are responsible for how you use the data and for respecting the authors' copyright; check WeChat's terms and your local rules.

Related actors

Building a wider China social-listening pipeline? Pair this actor with:

• Weibo Scraper — posts, profiles, comments and the realtime hot-search list from Weibo.
• Bilibili Video Scraper — videos, creators, comments and danmaku from China's largest video community.
• 1688 Scraper — products, suppliers and reviews from China's largest wholesale marketplace.