Datapika / Social & reviews
WeChat Scraper
Official Account articles, read counts and Channels. Article rows with title, summary, account, publish time and link; article details add full text with reads, likes, shares and saves; comment rows keep their replies.
Apify store response · 2026-10-30. Usage counts do not measure output accuracy.
· Build 1.0.8 · 5 rows reviewed. Previewing 5 below.
Swipe the table to inspect more fields →
| title | description | author_id | created_at | item_id |
|---|---|---|---|---|
| PHP大暴雷，全球程序员熬夜加班！超级漏洞还是ChatGPT发现的 | OpenAI：类似漏洞未来几个月数量会剧增。。。 | gh_114e76fd6e5d | 2026-10-21T16:43:34+00:00 | d9520aab0e9e825b29f709845cde1026 |
| 黄仁勋14天交付算力集群！配合DeepSeek开源引擎，单卡每小时38美金立刻开赚 | 算力也有自己的预制菜了 | gh_114e76fd6e5d | 2026-10-21T16:43:34+00:00 | 7556e8762881db6d90a49705b638bda5 |
| 多智能体协同重构医疗AI！清华AIR孵化：医生AI分身“坐诊”医院 | 医疗AI质变时刻 | gh_114e76fd6e5d | 2026-10-21T10:25:01+00:00 | 3e8ac87fc3ac386fd2fbd6cf32275cd6 |
| 南京物理AI上大分！三位图灵奖得主发声，50个应用场景亮相 | 让AI真正走进真实世界 | gh_114e76fd6e5d | 2026-10-21T10:25:01+00:00 | fc36f5115ecf4571f39a1da0c4076abd |
| LeCun押注的世界模型，中国一家清华系公司已经做进了机器人 | 解决机器人泛化，又有新方案了 | gh_114e76fd6e5d | 2026-10-21T06:24:33+00:00 | 109f6c50dadf6fc4a0499e98df384ddb |
Selected public fields from one recorded run. Text may be shortened to 20 words and non-HTTPS links are omitted. Missing values are preserved; this sample does not establish complete source coverage. Downloads contain all reviewed rows and selected fields. CSV text cells are escaped for spreadsheet safety.
Inspect the first JSON record
{
"title": "PHP大暴雷，全球程序员熬夜加班！超级漏洞还是ChatGPT发现的",
"description": "OpenAI：类似漏洞未来几个月数量会剧增。。。",
"author_id": "gh_114e76fd6e5d",
"created_at": "2026-10-21T16:43:34+00:00",
"item_id": "d9520aab0e9e825b29f709845cde1026",
"item_type": "article"
}02 / Run your own query
From a console, code or agent.
Use Apify's input form, or start a run with your own API token. The small starter input below was checked against build 1.0.8's published input schema on October 28, 2026. Recorded samples can use different inputs or older builds.
curl -X POST \
"https://api.apify.com/v2/acts/openclawai~wechat-scraper/runs?maxTotalChargeUsd=1" \
-H "Authorization: Bearer $APIFY_TOKEN" \
-H "Content-Type: application/json" \
-d '{"mode":"user_posts","urls":["gh_114e76fd6e5d"],"maxItems":10}'This starts an asynchronous paid run with a $1 total run cap. Use the returned defaultDatasetId to retrieve results after the run succeeds. Never put your API token in public frontend code.
03 / Field reference
Plan your data model.
Selected documented fields from build 1.0.8, reviewed October 28, 2026. Availability varies by mode and source. Missing and null values stay visible in the recorded samples.
| Field | Type | Meaning |
|---|---|---|
| item_type | string | Row type: article, comment, user, video, hint or error. |
| item_id | string | null | Article, comment or Channels video ID. |
| url | string | null | Article link on mp.weixin.qq.com. |
| title | string | null | Article or video title. |
| description | string | null | Article summary in lists; the full cleaned text in article details. |
| author_name | string | null | Official Account or Channels account name. |
| author_id | string | null | Account gh_ ID, or the Channels account ID. |
| created_at | string | null | Publish time in ISO 8601. |
| play_count | number | null | Reads in article details. |
| like_count | number | null | Likes in article details and on comments. |
| share_count | number | null | Shares in article details. |
| collect_count | number | null | Saves in article details. |
| comment_text | string | null | Comment text on comment rows. |
| cover_url | string | null | Cover image URL. |
04 / Know before you run
Coverage, without the guesswork.
Not an official WeChat API. WeChat answers in 5 to 15 seconds per request, account names are not unique (use the gh_ ID), and article details need a paid Apify plan.
Can I get WeChat article read counts?
Yes. The article details mode returns the cleaned full text with reads (阅读量), likes, shares and saves. In a run on October 10, 2026, a 量子位 (QbitAI) article returned 45,257 reads, 550 likes, 1,576 shares and 220 saves.
Why is WeChat slower than other platforms?
WeChat answers in 5 to 15 seconds per request. Account feeds return about 20 articles per request, article search about 40, Channels feeds about 15 videos and comment pages 15 to 25, so 1,000 articles take a few minutes.
The account name finds the wrong account. What now?
Names are not unique on WeChat. Use the gh_ ID or WeChat ID from the account search mode, or paste one of the account's article links.
How is the WeChat scraper priced?
Pay per event: one charge per article or Channels video from a list, comment, account from a search, or single lookup such as full article details, shown on the live price line above. Hint and error rows are free. Article details, profiles and comment replies need a paid Apify plan; the free plan allows 1 run per day with up to 5 results.