WeChat Articles Scraper — 微信公众号文章采集 (Search, Full Text & Monitoring)

Scrape WeChat Official Account (微信公众号) articles by keyword, by official account name, or from direct mp.weixin.qq.com links. Get the title, the real article URL, account name, account biz ID, author, exact publish time, digest, cover image and, optionally, the full text, HTML and images of every article. Monitoring mode returns only new articles, so you can schedule it and track topics, brands or competitors in China. No WeChat account, no login, no API key.

搜狗微信搜索关键词 / 公众号名称 / 文章链接，导出标题、链接、公众号、biz、作者、发布时间、摘要、封面、正文和图片。

What does WeChat Articles Scraper do?

• Keyword search of WeChat articles through Sogou WeChat search (weixin.sogou.com), up to 100 articles per keyword.
• Real article URLs: Sogou's /link?url=... redirects are resolved to https://mp.weixin.qq.com/s?... links.
• Official account names: returns indexed articles whose account name matches exactly (best effort, see FAQ).
• Direct article URLs: paste mp.weixin.qq.com/s/... links and get full content and metadata.
• Full content (optional): plain text, clean article HTML, all images, word count, author, account biz ID (__biz), account ID (gh_...), exact publish time and the original article (原创) flag.
• Only new articles mode for scheduled monitoring.
• Export as JSON, CSV, Excel, HTML, or use the Apify API, Make, Zapier, n8n, Google Sheets and webhooks.

Use cases

• China market and brand monitoring: track what WeChat publishers write about your brand, product or competitors.
• Media and PR tracking: collect coverage of a topic, event or company across official accounts.
• Investment and industry research: follow sectors such as 新能源, 芯片 or AI in Chinese-language media.
• AI and LLM datasets: full article text for RAG, summarization, translation and sentiment analysis.
• Competitor content analysis: see what accounts publish, how often, and whether it is original or reposted.

How to use it

1. Add keywords (Chinese works best, e.g. 人工智能), official account names (e.g. 36氪) and/or article URLs.
2. Set Max results per keyword/account (Sogou shows at most 100).
3. Turn on Include full article content for text, HTML, images, author and account IDs.
4. Click Start and download your data.

For monitoring, turn on Only new articles and schedule the Actor, for example every 6 hours. Each run checks the top Max results for each keyword and returns only the articles you have not received yet.

How much does it cost?

Pay per result, no subscription:

Event	Price
Actor start	$0.005 per run
article (one article from search or URL)	$0.004, or $4 per 1,000 articles
article-content (only with Include full article content, and always for direct URLs)	$0.006 per article

So search results cost $4 per 1,000 and articles with full text cost $10 per 1,000. Errors (blocked requests, deleted articles, invalid URLs) are never charged. Proxy costs are included.

Input

Field	Default	Description
keywords		Keywords to search on Sogou WeChat
accounts		Exact official account names (公众号名称)
articleUrls		https://mp.weixin.qq.com/s/... links
maxResults	20	Max articles per keyword and per account (1 to 100)
includeContent	false	Full text, HTML, images, author, biz ID, original flag
onlyNew	false	Return only articles not returned by previous runs
proxyConfiguration	Apify Proxy	Rotates IP and session automatically when Sogou shows a captcha

Example:

{
    "keywords": ["人工智能", "新能源汽车"],
    "accounts": ["36氪"],
    "articleUrls": ["https://mp.weixin.qq.com/s/_2kC-fXw7UjneZSrsC9CVQ"],
    "maxResults": 50,
    "includeContent": true
}

Output

{
    "title": "总台签约！讲好中国新能源汽车品牌故事",
    "url": "https://mp.weixin.qq.com/s?src=11&timestamp=1793064518&ver=7051&signature=ntmWgLLq...",
    "accountName": "央视财经",
    "publishTime": "2026-10-27T00:54:35.000Z",
    "digest": "△中央广播电视总台总经理室党委书记任学安与赛力斯集团董事、副总裁康波代表双方签约...",
    "coverImage": "https://mmbiz.qpic.cn/sz_mmbiz_jpg/nXgaYeEbuqia9Ac9iaDLd7fRRm5V.../0?wx_fmt=jpeg",
    "sogouDocId": "7931deb3a4f4979110ce9e9ec1c411f3",
    "source": "新能源汽车",
    "sourceType": "keyword",
    "searchQuery": "新能源汽车",
    "articleId": "MjM5NzQ5MTkyMA==_2658291567_2",
    "accountBiz": "MjM5NzQ5MTkyMA==",
    "accountId": "gh_9dc0e48d383a",
    "author": null,
    "isOriginal": false,
    "text": "10月26日，中央广播电视总台与赛力斯集团“品牌强国工程”战略合作签约活动在重庆举行……",
    "contentHtml": "<section>...</section>",
    "wordCount": 524,
    "images": ["https://mmbiz.qpic.cn/sz_mmbiz_jpg/y9A1qKVG.../640?wx_fmt=jpeg"],
    "permanentUrl": null,
    "scrapedAt": "2026-10-27T01:29:16.754Z"
}

accountBiz, accountId, author, isOriginal, text, contentHtml, wordCount, images and articleId (biz_mid_idx, stable across runs) are added by Include full article content. author is null when the publisher did not set one. Direct URL results also include permanentUrl (__biz, mid, idx, sn). Error rows have errorType and error.

FAQ

Do I need a WeChat account? No. The Actor reads only public pages: Sogou WeChat search and public mp.weixin.qq.com article pages.

Does it return read counts (阅读数) and likes (点赞/在看)? No. WeChat shows read, like and share counts only inside the WeChat app to a logged-in user, through a per-user key. They are not on the public article page, so this Actor does not return them, and it never invents them.

Are the article URLs permanent? Search results link to the real mp.weixin.qq.com article, but Sogou hands out signed links (src=11&timestamp=...&signature=...) that WeChat may expire after some time. Use articleId (with content on) as a stable key, and fetch content in the same run if you need the text. URLs you paste yourself stay as they are.

How many results per keyword? Sogou shows at most 100 articles (10 pages) per query to visitors who are not logged in. Use more specific keywords to get more coverage.

How do account names work? Sogou's account search no longer returns results, so the Actor searches articles with the account name and keeps only articles whose account name matches exactly. It finds recent articles of well-indexed accounts (e.g. 36氪), but it is not a complete account history.

What about captchas? Sogou shows a captcha (antispider) to an IP that sends too many requests. The Actor uses a fresh Apify Proxy IP and session for each block and retries, so you do not have to do anything. Datacenter proxy works; residential is not required.

How does "Only new articles" work? The Actor remembers the article IDs it returned for each keyword and account, in a key-value store in your account named wechat-articles-scraper-state. Delete that store to start over.

Is it legal? The Actor only reads public pages. You are responsible for how you use the data, including copyright and personal data rules.

Found a bug or need a field? Open an issue on the Actor's Issues tab. We usually reply within 24 hours.