WeChat Official Account Scraper — Articles & Search

Search public WeChat Official Account articles by keyword, or paste an mp.weixin.qq.com link and get the full text. No WeChat login, no browser.

Keyword search goes through Sogou's public WeChat index (weixin.sogou.com). Pasted article URLs are read from WeChat's own HTML (#js_content plus the account vars on the page).

Input	Row(s) emitted
Keyword (人工智能, Tesla)	~10 Sogou cards per page
Sogou listing URL (weixin.sogou.com/weixin?type=2&query=…)	Same cards
Article URL (mp.weixin.qq.com/s/…)	One full-text row

Pure HTTP. No browser, no WeChat account, no third-party captcha service.

Why Use This Scraper?

• Keyword search that returns a usable card. Title, Official Account name, date, snippet and cover, about 10 results per Sogou page.
• Full text when you already have the link. A public mp.weixin.qq.com/s/… URL returns body text, gh_ account id, __biz, cover and publish time.
• Only-new monitoring. Schedule a watch; IDs already delivered to this account are skipped and not billed.
• Named watches. Several schedules can keep separate seen-ID lists via monitorStateName.
• No login. Public pages only.

Overview

This actor is for people who need a structured stream of public Official Account articles — brand mentions, China news, competitor posts — without a WeChat account.

Every row is article-shaped. A keyword or a Sogou listing URL produces search cards (source: "sogou"). A pasted mp.weixin.qq.com URL produces a full article (source: "wechat") with contentText. Those two shapes share title / account / date / cover; only the pasted-URL path has body text and gh_ / biz.

Sogou does not put the real mp.weixin.qq.com URL on the listing. Following their /link?url= hop hits an anti-spider page, so search rows stay cards. Full text needs a real article URL.

This is the public-index actor. It does not return read counts, likes, comments, WeChat Index, or account registration data. Those live behind a paid WeChat data API (sian.agency sells them at $0.04–$1.50 a hit).

Supported Inputs

Input	Pattern	Example
Keyword	any Chinese or English topic	人工智能, Tesla
Sogou listing	https://weixin.sogou.com/weixin?type=2&query=…	article search, page N
Article URL	https://mp.weixin.qq.com/s/{id}	full text

Not supported

• WeChat Channels / 视频号
• Official Account profile search (type=1 on Sogou — the public HTML is empty)
• Private chats, Moments, login-gated articles
• Following a Sogou /link?url= hop (captcha)
• Read / like / comment counts

Use Cases

Who	What they do with the rows
Brand / comms	Watch a keyword daily and catch new Official Account mentions
China news / research	Pull titles, accounts and dates for a topic without a WeChat login
Competitor intel	Track what a known account is publishing, by pasting article URLs
Agencies	Feed a monitoring dashboard; only-new mode keeps the bill on fresh posts
LLM / RAG pipelines	Ingest contentText from pasted URLs, or titles + snippets from search

How It Works

1. Keywords become Sogou type=2 listing pages (?page=N, about 10 cards each).
2. Each card is written as a row: title, account, date, snippet, cover, Sogou link.
3. Pasted mp.weixin.qq.com URLs are fetched and parsed from the article page (og:title, nickname, user_name, biz, #js_content).
4. Optional monitor skips IDs already delivered to this account.
5. Output is one row per article, JSON or CSV.

Input Configuration

Field	Type	Required	Notes
keywords	string[]	no	Sogou article search. Example: 人工智能.
articleUrls	string[]	no	Direct WeChat article pages. Full text.
searchUrls	string[]	no	Pasted Sogou listing URLs (type=2).
maxItems	integer	no	Hard cap on emitted rows. Default 50. Free-tier runs cap at 100.
maxPages	integer	no	Sogou pages per keyword. Default 3, max 10.
monitorMode	boolean	no	Only-new. Skipped IDs are not billed. Default false.
monitorStateName	string	no	Name a watch so several schedules keep separate seen-ID lists.
emitBaseline	boolean	no	First monitoring run: true (default) emits current matches; false seeds silently.
includeUnchanged	boolean	no	Also re-emit already-seen rows, tagged changeType. Billed. Default false.
resetMonitoringState	boolean	no	Clear the seen-ID list at the start of this run.
proxy	object	no	Off by default. Sogou and public article pages answered direct in recon.

Example input

Keyword search plus one full article:

{
  "keywords": ["人工智能"],
  "articleUrls": ["https://mp.weixin.qq.com/s/urnBmheAWXNCDM5A8tAFYg"],
  "maxItems": 20,
  "maxPages": 2
}

Only-new daily watch:

{
  "keywords": ["Tesla"],
  "monitorMode": true,
  "monitorStateName": "tesla-daily",
  "maxItems": 50,
  "maxPages": 3
}

Output Overview

One dataset row per article. Search cards have source: "sogou" and a sogouUrl. Pasted pages have source: "wechat", articleUrl, accountId, biz and contentText. Monitoring adds changeType (new / unchanged).

Output Samples

Search card (source: "sogou"), from a live run:

{
  "type": "article",
  "source": "sogou",
  "title": "解读人工智能安全治理框架3.0",
  "accountName": "极客爱折腾",
  "snippet": "2026年9月14日,全国网络安全标准化技术委员会发布《人工智能安全治理框架3.0》…",
  "publishedAt": "2026-09-14T13:41:54.000Z",
  "coverUrl": "https://mmbiz.qpic.cn/…",
  "sogouUrl": "https://weixin.sogou.com/link?url=…"
}

Pasted article (source: "wechat"), from a live run:

{
  "type": "article",
  "source": "wechat",
  "title": "《麻省理工科技评论》万字长文：什么是人工智能？",
  "accountName": "麻省理工科技评论APP",
  "accountId": "gh_c4608d233c81",
  "biz": "MzI5MTcxMTA3Mw==",
  "authorName": "Will•D•Heaven",
  "publishedAt": "2024-07-14T14:03:50.000Z",
  "articleUrl": "https://mp.weixin.qq.com/s/urnBmheAWXNCDM5A8tAFYg",
  "contentText": "…"
}

Key Output Fields

Group	Fields
Identity	type, source, title, accountName
Account (pasted URL only)	accountId (gh_…), biz, authorName, avatarUrl
When	publishedAt, publishedTs
Body	snippet (always when present), contentText / contentHtml (pasted URL only)
Media	coverUrl
Links	articleUrl (pasted URL), sogouUrl (search card)
Monitor	changeType, scrapedAt

FAQ

Do I need a WeChat login? No. Public Sogou listings and public article pages only.

Why don't search rows have full text? Sogou hides the real article URL behind /link?url=, and that hop returns a captcha. Paste the mp.weixin.qq.com link when you have it.

Does this return read counts, likes or comments? No. Those are not on the public page. sian.agency sells them via a paid WeChat data API.

Can I search by Official Account name and get that account's archive? Not from the public Sogou type=1 HTML — it comes back empty. If you already have article URLs from that account, paste them.

Does this scrape WeChat Channels / 视频号? No. Official Account articles only. Channels would be a different actor.

Does monitoring charge for articles I already have? No. In monitorMode an ID already in your baseline is skipped and not billed. maxItems counts only rows that land in the dataset. The first run after you turn monitoring on returns the current matches and builds that baseline (set emitBaseline to false if you want the first run silent).

Can I run two watches from one account? Yes. Give each schedule its own monitorStateName so their seen-ID lists stay separate.

Do I need a proxy? Usually not. Enable one only if your IP is blocked.

Pricing

Event	When	Price
Actor start	Once per run	$0.005
Article row	Each card or full article written	$0.003

Skipped monitor IDs are not billed.

What this has that the other WeChat actors do not

	This actor	sian.agency	zen-studio	haketa / Sogou-only
Keyword cards (title, account, date, snippet, cover)	Yes	Paid API	Yes	Yes
Full text from a pasted mp.weixin.qq.com URL	Yes	Yes	Yes	Mixed
Only-new monitor + named watches	Yes — skipped IDs not billed	No	No	No
Price per row	$0.003	$0.04 search / $0.14 detail, up to $1.50/account	~$0.006	$0.000055–$0.01
Reads, likes, comments, WeChat Index	No	Yes (wxid-native API)	Mixed	No

If you need read counts, comments or account registration, use sian. That is a different product.

Support

Found a bug or need a field that isn't here? Open a ticket on the Issues tab.

Additional Services

Need this data pushed somewhere specific, or another China social surface covered? Get in touch through the Apify Console.

Explore More Scrapers

China social in this family: Douyin, RedNote / Xiaohongshu, RedNote search. Full list at apify.com/memo23.

🤖 For AI Agents & LLM Apps

Compact reference for agents calling this actor via the Apify MCP server or the Apify API (actor: memo23/wechat-official-account-scraper).

Purpose: public WeChat Official Account articles — Sogou keyword cards, or full text from pasted mp.weixin.qq.com URLs.

Minimal input:

{ "keywords": ["人工智能"], "maxItems": 10, "maxPages": 1 }

or

{ "articleUrls": ["https://mp.weixin.qq.com/s/urnBmheAWXNCDM5A8tAFYg"] }

Output: one dataset row per article — type, source, title, accountName, accountId, biz, authorName, snippet, publishedAt, publishedTs, coverUrl, avatarUrl, articleUrl, sogouUrl, contentText, contentHtml, scrapedAt, changeType.

Behaviors an agent should know:

• Always set maxItems. Default is 50; free-tier runs cap at 100.
• Search rows have no contentText and no articleUrl. Deduplicate on sogouUrl (cards) or articleUrl (pasted pages).
• monitorMode skips already-delivered IDs and does not bill them. Use monitorStateName for separate watches.
• Billing: $0.005 start + $0.003 per emitted row. Skipped monitor IDs are free.
• Unsupported: Channels / 视频号, Sogou account search (type=1), /link hops, reads/likes/comments.

⚠️ Disclaimer

This Actor is an independent tool and is not affiliated with, endorsed by, or sponsored by Tencent Holdings Limited (WeChat / Weixin / 微信), Sogou Inc., or any of their subsidiaries. All trademarks mentioned are the property of their respective owners.

The scraper accesses only publicly indexed Official Account articles and Sogou's public WeChat search pages — no authenticated endpoints, paid WeChat data APIs, private chats, Moments, or content behind a login wall. It does not follow Sogou /link hops that return an anti-spider page. Users are responsible for ensuring their use complies with WeChat and Sogou terms of service, applicable data-protection law (GDPR, CCPA, and PRC personal-information rules), and any contractual obligations of their own organization.

SEO Keywords

wechat scraper, weixin scraper, wechat official account scraper, 公众号 scraper, 微信公众号, sogou wechat, mp.weixin.qq.com scraper, wechat article api, wechat article search, weixin article scraper, china social scraper, wechat brand monitoring, wechat news scraper, official account monitor, china content intelligence, wechat competitor research, sogou weixin search, wechat public account api, 微信文章 scraper, wechat media monitoring