WeChat scraper API for official account articles
Send a public WeChat article URL to the Crawl endpoint and get back clean Markdown — the article body, account name, publish time, and image URLs. No WeChat account needed.
Endpoint — bash
A dedicated parser extracts the article body — the content WeChat serves on the public article page.
The full article as clean Markdown — headings, lists, and formatting preserved from #js_content.
The official account name and publish timestamp ride along in structured metadata.
Every image in the article is listed in metadata.imageUrls — no inline base64 blobs.
Useful for
Chinese content RAG / Account monitoring / Archive pipelines
What goes in, what comes back
One real public article. Left: the page as WeChat actually serves it to anonymous readers — a real capture through the same egress the crawler uses. Right: the complete results.content the API returns, byte for byte.
mp.weixin.qq.com/s/jUylk813LYbKw0sLiIttTQWhat WeChat serves
5.1 MB of rendered HTML — ad modules, QR cards, related reads
What the API returns
markdown
Full article — 14 KB · account + publishedTime + 12 image URLs
What is covered, by URL pattern
WeChat support is measured per URL pattern, not claimed per domain. Every row below is verified against the live crawler.
| URL pattern | Status | What you get |
|---|---|---|
mp.weixin.qq.com/s/<id> | Supported, egress-dependent | Public articles return the full body, account, publish time, and image URLs — but Tencent occasionally challenges some egress IPs with a captcha (see below). |
captcha-challenged requests | Named error | When Tencent serves its anti-bot page, the request fails as an ANTI_BOT-classified error instead of returning captcha markup — retry rotates egress. |
Official account articles, minus the wrapper
WeChat article pages are heavy — the piece you want sits inside megabytes of account chrome, ad modules, and related-read widgets. Point the Crawl endpoint at an mp.weixin.qq.com URL and a dedicated parser extracts the article body plus account name, publish time, and image URLs.
The body, not the wrapper
The parser targets the article container directly — the account card, ad slots, and share chrome never reach your pipeline.
Images as URLs
Article images come back in metadata.imageUrls — referenceable links, not inline blobs.
Anti-bot is named, not hidden
Tencent occasionally serves a captcha page to datacenter egress — those requests fail with an ANTI_BOT classification you can catch and retry, never silently returning challenge markup.
Verified per URL pattern
Every row in the matrix is reproduced from a live crawl of a real URL — support is measured per pattern, not claimed per domain.
What is a WeChat scraper API?
A WeChat scraper API is an HTTP interface that takes an mp.weixin.qq.com article URL and returns the article in a structured, model-ready format. Public-account articles are one of the largest bodies of Chinese-language writing on the web, but the page is a dense wrapper around a single content container. Search1API’s Crawl endpoint uses a dedicated parser that extracts that container — body, account, publish time, and image list — and reports honestly when Tencent’s anti-bot answers instead of the article.
Typical workflow
Send a public WeChat article URL to the Crawl endpoint and get back clean Markdown — the article body, account name, publish time, and image URLs. No WeChat account needed.
POST an mp.weixin.qq.com article URL to the Crawl endpoint — no WeChat account needed.
The URL is matched against the pattern rules and routed to the dedicated WeChat parser.
The article returns as Markdown plus results.metadata — account, publishedTime, imageUrls.
If Tencent challenges the request, the error carries an ANTI_BOT classification — catch it and retry instead of ingesting a captcha page.
Where teams use it
Feed Chinese-language articles into RAG pipelines — the body arrives clean, with images as URL lists.
Monitor official accounts — the account name and publish timestamp come back as fields.
Archive public-account writing without reverse-engineering the page wrapper per article.
Build reading lists — an article URL returns the text, not the megabyte of chrome around it.
FAQ
Do I need a WeChat account?
No — public official-account articles are served to anonymous browsers, and the parser reads the same page. Anything that requires the app or a login stays out of scope.
Which WeChat URLs are supported?
Public article URLs of the form mp.weixin.qq.com/s/<id> — the standard share link for official-account posts.
Sometimes the same article fails — why?
Tencent occasionally serves an anti-bot captcha page to datacenter egress IPs. When that happens the request fails with an ANTI_BOT-classified error rather than returning challenge markup — retry rotates the exit and usually succeeds. The matrix documents this honestly.
What metadata comes back?
The official account name, the publish timestamp, and the full list of image URLs in the article — platform is "weixin".
Do the article images come back inline?
No — images are listed in metadata.imageUrls as links. The Markdown body carries the text; the image list carries the assets.
Is scraping WeChat legal?
We only fetch article pages WeChat serves to anonymous visitors — no login, no session sharing, no circumvention of access controls. Whether a specific use complies with applicable law and Tencent’s terms depends on your jurisdiction and use case, so confirm that with your own counsel.