WeChat Content Studio

The Problem

Writing a WeChat Official Account article often breaks down into manual steps: finding sources, comparing articles, extracting body text, rewriting, adding a cover, and preparing publishing. When material comes from WeChat articles, Zhihu, CSDN, Juejin, GitHub, or arXiv, the editor has to switch between sites, clean up inconsistent text, and re-organize references. This skill turns that path into a local workflow: retrieve sources, extract articles, rewrite them, generate a cover, and prepare output for publishing.

How It Works

The workflow is organized around commands such as search, links, and workflow. Multi-source search covers WeChat Official Accounts, Zhihu, CSDN, Juejin, GitHub Trending, arXiv, and Hacker News, with category keys like high_quality_channels, wechat_top, and dev_community. Link extraction and multi-site extraction support WeChat, CSDN, Cnblogs, Juejin, Zhihu, Jianshu, Sspai, and general web pages, so scattered materials can be consolidated into one working directory.

In merged output, the result is written under a sanitized article-title folder containing article.md; in non-merged mode, files are split into 1_title/, 2_title/, and so on, which helps keep multiple sources separate. The rewriter adjusts tone, removes redundant phrasing, and reduces formulaic AI-like expression. Cover generation uses Alibaba Cloud Wanxiang 2.6, with fallback reuse of a previously uploaded cover. Publishing can use wenyan-cli or browser-use, while layout depends on wechat-typeset-pro. Configuration is usually loaded from ~/.openclaw/.env or the skill directory, including variables such as WECHAT_APP_ID, WECHAT_APP_SECRET, DASHSCOPE_API_KEY, and OPENAI_API_KEY.

Boundaries and Notes

This is useful for assembling research, extracting articles, and producing a first draft, but the output still needs human review for facts, tone, and copyright. Publishing requires WeChat API credentials and IP whitelisting in the official account backend. Heavy use of Brave Search may hit HTTP 429; reduce --count, avoid broad source sweeps, or increase BRAVE_SEARCH_MIN_INTERVAL_MS. Because the workflow consumes external content, keep attribution where appropriate and rewrite deeply rather than copying source articles directly.