← ClaudeAtlas

zhihu-publishlisted

Use when the task is to publish a Zhihu article from Markdown using the local browser-automation pipeline in this repo. Covers the text-only publishing chain, the multimodal chain that uploads Markdown images into the Zhihu editor, cookie-based login checks, and choosing between the stable text script and the richer multimodal script.
Harzva/learn-likecc · ★ 61 · AI & Automation · score 75
Install: claude install-skill Harzva/learn-likecc
# Zhihu Publish ## Overview This skill uses the repo-local Zhihu publishing scripts under `wemedia/zhihu/` to publish Markdown articles through Zhihu's web editor. Use it when the user wants to: - publish a Markdown article to Zhihu - verify whether `cookies.json` still works - use the text-only chain first - try the multimodal chain for Markdown files that contain local images - overwrite an existing Zhihu article instead of creating a new one - include generated article visuals from HTML/CSS flow cards or webpage screenshots ## Workflow ### 1. Choose the pipeline Prefer these paths: - text only or first-pass stability: `wemedia/zhihu/publish_article_text.js` - text + Markdown images: `wemedia/zhihu/publish_article_multimodal.js` - overwrite an existing article: `wemedia/zhihu/edit_article_multimodal.js` Keep `wemedia/zhihu/publish_article.js` as the general entry if the repo later repoints it, but when reliability matters, call the explicit script. ### 2. Confirm prerequisites Before publishing: - `wemedia/zhihu/cookies.json` must exist - the target Markdown article must exist - image files referenced by Markdown must exist if using the multimodal path Useful checks: ```bash ls -l wemedia/zhihu/cookies.json ls -l wemedia/zhihu/articles/your-article.md ``` If the article needs images first, generate them before publishing: - webpage section screenshots: use `webpage-screenshot-md` - small explainer graphics from HTML/CSS: use `html-card-to-png` ### 3. Run th