WeChat Web Publisher

Problem

Turning a web page or local document into a WeChat article usually requires manual copy-paste, image handling, layout adjustment, and dealing with anti-bot pages, drafts, and publishing. Web Publisher moves fetching, document conversion, optional rewriting, drafting, and publishing to the server side and exposes them through HTTP calls.

How It Works

• Login and setup: login returns verifyUrl and userCode; after the user confirms in the browser, login-status pulls the credentials. wechat config returns a setup URL and requires adding serverIps to the WeChat official-account IP whitelist.
• Draft and publish: draft accepts an http(s):// URL or a local document path such as PDF, DOCX, PPTX, XLSX, or EPUB; publish is used only when the user explicitly asks.
• Convert and rewrite: convert can output Markdown only; for large files, use --out to write to disk. --rewrite can add a rewrite style.

Limits and Caveats

• Draft is the default; do not publish proactively, and avoid timeout retries that may create duplicate drafts.
• Local files are capped at 50 MiB; file:// is not supported. For academic PDFs or large files, prefer --async.
• Server-side fetching uses fixed outbound IPs, so some Xiaohongshu, Zhihu, login-walled, or overseas pages may return 403 or login walls; a local fallback may be needed.
• AppSecret never enters the conversation context and is submitted only through the browser form.