WeChat Articles Scraper

WeChat articles scraper: search Official Account articles by keyword or brand name and get one row per article: title, summary, account, WeChat ID, verified flag, publish time and cover, and with the full text on, the body, author, publishing region, tags and images. It also finds the accounts writing about a topic, lists one account's articles, reads every article in an album, and opens article links you already have. A keyword reaches up to 100 articles, as deep as the public search goes without signing in, so a brand watch works best with a few variants of the name.

Input	Keywords or brand names, account names or WeChat IDs, article links, album links
Output	One row per article: title, summary, account, WeChat ID, verified, publish time, cover, links. With the full text: body, author, gh_ ID, region, tags, original badge, album, images
Ceiling	100 articles a keyword or account, 1,000 an album, 10,000 rows a run
Speed	About 300 to 500 articles a minute from search, about 15 to 25 a minute with the full text (real runs, 3 November 2026)
Account needed	None from you
Price	$0.0145 per 1,000 articles, the same on every plan. The full text adds $0.80 per 1,000, accounts are $0.205 per 1,000 and converted links $1.25 per 1,000. The free plan's $5 a month covers about 340,000 articles, or about 6,000 with the full text

🌏 中文说明

按关键词或品牌名搜索微信公众号文章，每篇文章一行：标题、摘要、公众号名称、微信号、是否认证、发布时间和封面。开启全文后，还有正文、作者、原始ID、发布地区、话题标签、原创标记、所属合集和图片。也可以查找写某个话题的公众号、某个公众号的文章、合集里的全部文章，或读取你已有的文章链接。

只需填写"关键词"一栏即可开始。按条数收费，价格见上方表格。不需要账号。本工具不是微信官方产品。

🔍 What WeChat Articles Scraper does

Pick one job in What to get:

• Articles that mention my keywords: the articles a search for each keyword finds, newest first if you like, filtered by date, by account or by words to leave out.
• Accounts behind those articles: one row per account writing about the keyword, with how many matching articles it has and its latest one.
• One account's articles: give a name, a WeChat ID or a link to one of its articles.
• Article details from links: the full text and details of mp.weixin.qq.com links you already have.
• Every article in an album: a 合集, newest first, each with its permanent link.
• Convert article links: the short link, the article's ids and, where WeChat gives it, the permanent link.

With Only new since my last run on, a daily schedule returns only what earlier runs did not, and you pay for those alone.

📋 What data you get from each WeChat article

You want	Fields
The article	title, summary, publishedAt (Beijing time), publishTimestamp, coverUrl, hasVideo, readMoreUrl
The account	accountName, accountWeChatId, accountVerified, accountAvatarUrl, accountOpenId
Where it was found	query, page, position, source, searchSignals
Links and ids	searchResultUrl, articleUrl, articleKey, messageId, indexInMessage
Full text (paid add-on)	contentText, characterCount, author, accountId, accountDescription, isOriginal, isPaid, ipRegion, hashtags, topics, album, itemShowType, commentId, biz, images, imageCount, links, linkCount, videoIds, videoCount
Optional with the full text	contentMarkdown, contentHtml
Album rows	albumTitle, albumId, albumArticleCount, albumReadCount, permanentUrl, sn
Account rows	articlesFound, latestArticle, plus accountId and accountDescription with the full text on

▶️ How to scrape WeChat articles

1. Open the actor and click Try for free.
2. Leave What to get on "Articles that mention my keywords" and type your words in Keywords, one per line, in Chinese or English.
3. Set Results per keyword, account or album to how many you want, up to 100.
4. Turn on Add the full text (paid add-on) if you need the body, or set Published within for recent articles only.
5. Click Start, then export the dataset as JSON, CSV or Excel.

💰 How much does it cost to scrape WeChat articles?

An article row costs $0.0145 per 1,000, and every plan pays the same. The full text adds $0.80 per 1,000 articles whose page was read. An account row is $0.205 per 1,000, and a converted link, or the article link added to a search row, is $1.25 per 1,000. A run also has a start fee of $0.00005. The free plan's $5 a month covers about 340,000 articles, or about 6,000 with the full text.

Not charged: articles left out by your filters, the same article found twice in a run, articles returned before when Only new since my last run is on, and the note rows that say why something is missing.

📥 What you give it

Field	Default	What it does
mode	search	The job: search, accounts, accountArticles, articles, albums or links
keywords	none	Words or brand names, for search and accounts
searchUrls	none	Links to search result pages you already have; a date filter in them is kept
accounts	none	For accountArticles: the name as WeChat shows it, the WeChat ID, the gh_ ID or an article link
articleUrls	none	For articles and links: mp.weixin.qq.com links, short or long, or search result links
albumUrls	none	For albums: links with __biz and album_id
maxResultsPerInput	20	Rows per keyword, account or album; a keyword or account stops at 100, an album at 1,000
maxItems	100	Rows in the whole run, up to 10,000
sortBy	relevance	relevance, or newest: latest first. A keyword search then starts with the last day's articles, and the more rows you ask for, the deeper it reads, up to all 100 results
publishedWithin	any	day, week, month, halfYear or year
publishedAfter	none	A date like 2026-10-01 (Beijing time) or a number of days like 7; other forms are refused before anything runs
publishedBefore	none	A date like 2026-10-31
excludeKeywords	none	Leave out articles whose title or summary has any of these words
onlyFromAccounts	none	Keep only articles by these accounts, by name or WeChat ID
includeContent	false	The full text and details, a paid add-on
includeMarkdown	false	Also the text as Markdown
includeHtml	false	Also the text as HTML
includeImages	true	The list of images in the body
resolveArticleUrls	false	Add the article's own link to search rows, charged as a link
onlyNew	false	Leave out what earlier runs with the same monitor name returned
monitorName	wechat-articles-seen	Letters, digits and hyphens; one name per schedule
resetMonitor	false	Start the monitor's list again
requestDelaySecs	1	Seconds between two search pages, 0.5 to 10
proxyUrls	none	Optional. Servers of your own you want the run to use, as http://user:pass@host:port

{
  "mode": "search",
  "keywords": ["比亚迪", "BYD"],
  "maxResultsPerInput": 50,
  "publishedWithin": "month",
  "excludeKeywords": ["招聘"],
  "includeContent": true
}

📤 What you get back

One row from a real run on 3 November 2026, trimmed:

{
  "recordType": "article",
  "source": "search",
  "query": "华为",
  "page": 1,
  "position": 5,
  "articleKey": "oIWsFtwbC_RfPo29wSlbi2WbTTF4-2650876517-2",
  "title": "华为最新回应",
  "summary": "昨日(6月16日)晚间,华为就美国商务部发布的新规做出回应...",
  "accountName": "南方都市报",
  "accountWeChatId": "nddaily",
  "accountVerified": true,
  "publishedAt": "2020-06-17T12:15:17+08:00",
  "publishTimestamp": 1592367317,
  "coverUrl": "http://mmbiz.qpic.cn/mmbiz_jpg/3OEpTPib0kV9Br1fV...",
  "searchResultUrl": "https://weixin.sogou.com/link?url=dn9a_-gY295K0Rci_xozVXfd...",
  "articleUrl": "https://mp.weixin.qq.com/s?src=11&timestamp=1793690568&ver=7065&...",
  "textStatus": "read",
  "accountId": "wxid_4783107810411",
  "accountDescription": "南都，换一种方式",
  "isOriginal": false,
  "ipRegion": { "country": "中国", "province": "广东" },
  "album": null,
  "biz": "MTk1MjIwODAwMQ==",
  "contentText": "昨日（6月16日）晚间，华为就美国商务部发布的新规做出回应。华为官方声明...",
  "characterCount": 1046,
  "imageCount": 2,
  "linkCount": 6
}

Field	How to read it
articleUrl	On a search row it is signed for that search and opens the article for some hours. Album rows and the links you give carry links that last
articleKey	The same for an article on every run: join and deduplicate on it
accountId	Usually a gh_ ID; older accounts can show a wxid_ form
ipRegion, album, author	null when the article page shows none
searchSignals	The ranking figures the search gives the result, passed on as given
publishedAt	Beijing time, with its +08:00 offset

🧾 Reading the output

• recordType is article, account or link. Every one of those rows is charged.
• With the full text on, textStatus says read, gone (WeChat no longer shows the article, and textMessage gives its reason) or not_served (the page was not served this run). Only read rows pay the add-on.
• A note row has _diagnostic: true and ok: false, costs nothing, and its status is one of: BAD_INPUT (the input was refused before anything ran), NO_RESULTS (nothing for that keyword or album, or nothing that passed your filters), NOT_FOUND (the link or account does not exist), NOT_SERVED (it was not served this run; run it again later), TIME_LIMIT, COST_LIMIT (your run limit or cost limit stopped it), UNEXPECTED_ERROR (with a reference to quote).
• To keep only paid rows, drop the ones where _diagnostic is true.

💡 What people use it for

• A brand team in China checking every morning what Official Accounts wrote about the brand overnight, with Only new since my last run on.
• A PR agency collecting launch coverage for a client report, full text included, filtered to the week after the launch.
• An analyst building a reading list of an industry, starting from the accounts that write about it most.
• An editor archiving a column that lives in one WeChat album.

A recipe: run Accounts behind those articles for your category, take the accountName values of the accounts that matter, then schedule One account's articles on them daily with only-new on. For the same brand on Weibo, run Weibo Scraper on the same words.

🚧 What it does not do

• It returns image links, not image or video files.
• It reads public articles as a visitor sees them; it does not post, comment or follow.
• It is not an official WeChat tool. Dami's Studio is independent and is not affiliated with or endorsed by WeChat, or by any other company named on this page.

🧭 Which WeChat scraper do you need?

If you want	Use
WeChat articles about a brand or topic, the accounts writing them, one account, an album, or links you have	This one
The same brand in Weibo posts, with the authors' follower counts	Weibo Scraper
What people search for on Douyin right now	Douyin Hot Search Scraper
RedNote notes you have links for, with their like and collect counts	RedNote Scraper
Bilibili videos, rankings and the uploaders behind them	Bilibili Scraper

❓ Questions people ask

Do I need a WeChat account to scrape WeChat articles?

No. You don't need a WeChat account or any key. Articles are read the way anyone opening them in a browser sees them.

How do I cover more than 100 articles for one brand?

Add variants as separate keywords: the Chinese name, the English one, product lines, the name of the brand's own account. Then use Accounts behind those articles and One account's articles on the accounts that matter, and schedule the run with only-new on so each day adds what is new.

Why does an article link from a search stop opening?

The link on a search row is signed for that search and opens for some hours. Keep articleKey, the title and the account to find the article again, and search again when you need a link that opens. Album rows and the links you give carry links that last.

Can I call it from code or an AI assistant?

Yes. The actor's API tab has ready calls in several languages. For an MCP client, use https://mcp.apify.com/?tools=dami_studio/wechat-articles-scraper, and set maxTotalChargeUsd on the run to cap the bill.

Is scraping WeChat legal?

The articles are public, and this actor reads them the way a signed-out visitor sees them, with no account and nothing private. Personal data in them falls under GDPR and similar laws, so keep only what you need. WeChat's own terms still apply to what you do with the data. Apify has a write-up on the legality of web scraping. We are not lawyers.

🆘 If something breaks

Open an issue on the actor's Issues tab with the run's id and the input you used. Note rows in the dataset usually say what happened first.