WeChat Account Articles Scraper

✨ What you get

EN — Fetch a paginated article list from any WeChat Official Account (公众号). Returns title, digest, cover, content URL and create_time for every article. Supports auto-pagination using the data provider base64 next_offset cursor, with a configurable page cap.

ZH — 抓取任意微信公众号的文章列表。返回标题、摘要、封面、正文链接、发布时间。支持 base64 next_offset cursor 自动翻页,翻页页数可配置。

---

💼 Use cases

• Inventorying everything one account has published — point username at a gh_… ID and the Actor walks the account's own listing, returning title, digest, cover, url and createTime per row, so you get the catalog instead of the couple of posts a share dialog exposes.
• Watching a competitor or KOL for new posts — schedule one run per day per account, as the README suggests, and diff articleId sets between runs to see exactly what shipped since yesterday.
• Covering accounts that post in more than one format — itemShowType switches the listing between articles (0), videos (5), audios (7) and image-text posts (8), so a mixed-format account does not need a second tool.
• Feeding clean article URLs into the next step — each row's url plus the covers variants (16:9, 1:1, 235:1, 3:4) drop straight into the detail, comments and metrics Actors, or into your own thumbnail slot.

✅ Who this is for

• Content and editorial teams — who track a stable set of accounts and want one row per post in a schema that loads into a spreadsheet or BI tool.
• KOL and partner vetting — where the first question is how much this account actually publishes before anyone signs.
• Competitor monitoring programmes — that need a repeatable, cursor-paginated listing instead of manual copying out of the WeChat app.
• Chinese-market SEO and content repurposing — work, where title, digest and publish timestamp are the raw material for an editorial calendar.

❌ Who this is NOT for

• You need a fixed page size — itemsPerPage is forwarded to the platform but WeChat currently ignores it, and the real count per page is decided by the account, so page length is not yours to control.
• You only have an account name — the input is the gh_… ID sitting in the __biz parameter of any article link (or a custom micro-signal), never a display name. Start from a keyword with the universal search Actor, then feed the username back here.
• You want bodies, engagement numbers or comments — this listing is metadata only: no contentText, no read / like / share counts, no comment text. Each of those is a separate call per URL.
• You are cost-scanning a deep back catalogue — billing is per page fetched, so maxPages=10 is ten billable events and the cap is fifty pages. Budget in pages, not posts, and remember free-tier runs are limited per day.

🚀 Quick start

1. Console: paste any gh_… username (or use the default gh_114e76fd6e5d) and click Start.
2. API: apify call wechat-account-articles-scraper -i '{"username":"gh_114e76fd6e5d","maxPages":2}'
3. Schedules: schedule once per day per account to monitor new posts.

Free users: 5 runs/day. Paid users: charged 1 PPE event per page fetched (account-articles-result).

---

📥 Input parameters

Field	Type	Default	Notes
username	string	gh_114e76fd6e5d	Three forms: gh_<hex>, gh_<hex>@app, or a custom micro-signal (e.g. nikejdi).
maxPages	int	1	Auto-pagination cap (1–50). Stops early when Data provider returns is_end=1.
itemsPerPage	int	20	Page size hint forwarded to Data provider. NOTE: WeChat currently ignores this — actual count is decided by the account itself.
itemShowType	string	"0"	0 = articles (default), 5 = videos, 7 = audios, 8 = image-text posts. Empty string falls back to default.
raw	string	"false"	false = omit raw Data provider payload (smaller dataset); true = attach rawData[] (one entry per page).

How to find the username

You usually get a username from an article URL. The user_name is the __biz query parameter in any https://mp.weixin.qq.com/s/... link. For example:

https://mp.weixin.qq.com/s?__biz=MzIzNjc1NzUzMw==&mid=2247923517&idx=1&sn=...
                                       ^^^^^^^^^^^^^^^^^^^^^^
                                       this base64 decodes to the gh_<hex> form

Or use the WeChat Article Detail Scraper — the output's userName field is exactly the value you pass into this Actor.

---

📤 Output

{
  "username": "gh_114e76fd6e5d",
  "bizUsername": "gh_114e76fd6e5d",
  "hasMore": false,
  "nextOffset": "",
  "totalFetched": 23,
  "pagesFetched": 2,
  "itemShowType": 0,
  "articles": [
    {
      "articleId": "2247923517",
      "idx": 1,
      "title": "被英伟达点名的杭州团队，补上了AI for Science的「最后一公里」",
      "digest": "从想法到结果，一次对话实现",
      "url": "http://mp.weixin.qq.com/s?__biz=...&mid=2247923517&idx=1&sn=...",
      "cover": "https://mmbiz.qpic.cn/.../640?wxtype=jpeg",
      "covers": {
        "ratio16_9": "https://mmbiz.qpic.cn/.../640?wxtype=jpeg",
        "ratio1_1": "https://mmbiz.qpic.cn/.../300?wxtype=jpeg",
        "ratio235_1": "https://mmbiz.qpic.cn/.../300?wxtype=jpeg",
        "ratio3_4": "https://mmbiz.qpic.cn/.../540?wxtype=jpeg"
      },
      "createTime": 1789519199,
      "updateTime": 1789526559,
      "isPaid": 0,
      "isPaySubscribe": 0,
      "itemShowType": 0,
      "itemShowTypeLabel": "article",
      "msgType": 9,
      "picCount": 0,
      "sourceUrl": null
    }
  ],
  "collectedAt": "2026-09-16T12:00:00.000Z"
}

Field notes

• articleId is emitted as a string (per §98 / large integer precision).
• createTime / updateTime are Unix epoch seconds.
• covers includes 16:9, 1:1, 235:1 and 3:4 aspect variants from Data provider; pick the one that matches your layout.
• hasMore=false and nextOffset="" means the data provider has no more articles; the Actor stopped automatically.
• When raw=true, each rawData[i] is the data provider data object for page i+1.

---

💰 Billing

Charged once per page fetched. Auto-pagination is transparent — maxPages=10 means 10 PPE events. Stopping early on is_end=1 does not charge the unused budget.

---

📚 Resources

$npm test

Runs unit assertions (input validation, body shape, normalize across raw=true / raw=false, simplified / raw article shapes, multi-page aggregation) plus 9 integration scenarios (mock fetch drives the full run() flow including pagination, free-tier gating, Data provider errors).

✅ unit.js: all assertions passed.
integration.js: 9 scenarios passed.

📜 Notes & caveats

Use it together with the other WeChat Actors

Where in the pipeline	Actor	Console	Use it to
step ① profile	Account Profile Scraper	…/3JV499ECEjb5un854	step ① — fetch the account identity + service type before listing its articles
step ② listing (you are here)	Account Articles Scraper	…/actor	step ② — list every article published by a given account
step ③ detail	Article Detail Scraper	…/5vrr8dBcnclIva0Lx	step ③ — pull the full body + author + publish time for each article URL
search-first	Universal Search Scraper	…/search	sister Actor — discover accounts by keyword, then feed their userName back into this Actor

You are here: Account Articles Scraper (step ②) — run it to inventory everything an account has published.

---