wechat-article-extraction

スター12

フォーク2

更新日2026年7月6日 13:33

Extract WeChat (mp.weixin.qq.com) articles to Markdown with images, headings, lists, and formatting preserved.

インストール

Codex または Claude でインストールこの Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。

ソース

werxiao882-cell

werxiao882-cell/NanoHermes

GitHub リポジトリを開く Creator のリポジトリを見る

ダウンロード

Overview

Extract articles from mp.weixin.qq.com/s/<id> and convert to clean Markdown with images, headings, lists, and tables preserved.

Extraction Method

Download HTML via curl (browser is unreliable due to captcha):

curl -s -L -A "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36" "https://mp.weixin.qq.com/s/<article_id>" > /tmp/wechat_article.html

Extract article content using Python + BeautifulSoup (run via terminal with conda Python, NOT execute_code sandbox):
- Locate id="js_content" div
- Walk the DOM tree collecting text nodes, images (<img> with data-src), and styled <section> elements
- Detect headings via font-weight: bold or font-size: 16px+ inline styles
- Convert **bold**, *italic*, lists, blockquotes to Markdown
- Extract background images from section style="url(...)" declarations
- Preserve image alt text as Markdown image alt attributes
Image placement: Process images in document order — each <img> is placed at its natural position relative to surrounding text, NOT batched at the top.
Cleanup: Remove JavaScript artifacts (var first_sceen__time, etc.), stray UI text (预览时标签不可点), and excessive blank lines.

Platform-Specific Notes

WeChat (mp.weixin.qq.com): ✅ Works reliably via curl + BeautifulSoup. No login needed. Anti-bot is mild (captcha only in browser).
Zhihu (zhuanlan.zhihu.com): ❌ Fails consistently. Returns 40362 error. API returns code 10003. The zse-ck anti-bot blocks all non-browser requests. Workaround: ask user to provide content manually, or search cached/mirrored versions.

Pitfalls

Images lost during extraction: WeChat wraps images in <section> > <figure> > <span> > <img>. Must recurse into sections and handle <img> before recursing into text children.
Headings missed: WeChat uses inline styles (font-weight, font-size) rather than <h1>-<h6> tags. Must check style attributes, not just tag names.
Execute_code sandbox lacks packages: bs4 (BeautifulSoup) is NOT available in the execute_code sandbox. Use terminal with conda-activated Python instead.
f-string backslash error: Python f-strings cannot contain backslashes in the expression part. Use a variable for regex patterns instead.

Reference Files

references/vocabulary-glossary-guide.md — How to create plain-language glossaries for technical articles
references/hermes-context-explanation-guide.md — How to explain abstract concepts using the user's own project as analogy

このリポジトリの他の Skills

同じリポジトリ

ai-agent-integration-testing

werxiao882-cell/NanoHermes

Integration testing for AI Agent CLI systems — PTY-driven conversation simulation, tool chain verification, OpenSpec-driven test design, and real-API validation.

2026-07-0612

article-extraction

werxiao882-cell/NanoHermes

Extract web article content from WeChat MP, blogs, and other platforms, converting text, images, formatting, and tables to clean Markdown.

2026-07-0612

hermes-agent

werxiao882-cell/NanoHermes

Configure, extend, or contribute to Hermes Agent.

2026-07-0612

comfyui

werxiao882-cell/NanoHermes

Generate images, video, and audio with ComfyUI — install, launch, manage nodes/models, run workflows with parameter injection. Uses the official comfy-cli for lifecycle and direct REST/WebSocket API for execution.

2026-07-0612

kanban-orchestrator

werxiao882-cell/NanoHermes

Decomposition playbook + anti-temptation rules for an orchestrator profile routing work through Kanban. The "don't do the work yourself" rule and the basic lifecycle are auto-injected into every kanban worker's system prompt; this skill is the deeper playbook when you're specifically playing the orchestrator role.

2026-07-0612

kanban-worker

werxiao882-cell/NanoHermes

Pitfalls, examples, and edge cases for Hermes Kanban workers. The lifecycle itself is auto-injected into every worker's system prompt as KANBAN_GUIDANCE (from agent/prompt_builder.py); this skill is what you load when you want deeper detail on specific scenarios.

2026-07-0612

name	wechat-article-extraction
category	research
description	Extract WeChat (mp.weixin.qq.com) articles to Markdown with images, headings, lists, and formatting preserved.
trigger	Extract content from WeChat (微信公众号/mp.weixin.qq.com) articles, convert to Markdown, preserve images and formatting.

Overview

Extract articles from mp.weixin.qq.com/s/<id> and convert to clean Markdown with images, headings, lists, and tables preserved.

Extraction Method

Download HTML via curl (browser is unreliable due to captcha):

curl -s -L -A "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36" "https://mp.weixin.qq.com/s/<article_id>" > /tmp/wechat_article.html

Extract article content using Python + BeautifulSoup (run via terminal with conda Python, NOT execute_code sandbox):
- Locate id="js_content" div
- Walk the DOM tree collecting text nodes, images (<img> with data-src), and styled <section> elements
- Detect headings via font-weight: bold or font-size: 16px+ inline styles
- Convert **bold**, *italic*, lists, blockquotes to Markdown
- Extract background images from section style="url(...)" declarations
- Preserve image alt text as Markdown image alt attributes
Image placement: Process images in document order — each <img> is placed at its natural position relative to surrounding text, NOT batched at the top.
Cleanup: Remove JavaScript artifacts (var first_sceen__time, etc.), stray UI text (预览时标签不可点), and excessive blank lines.

Platform-Specific Notes

WeChat (mp.weixin.qq.com): ✅ Works reliably via curl + BeautifulSoup. No login needed. Anti-bot is mild (captcha only in browser).
Zhihu (zhuanlan.zhihu.com): ❌ Fails consistently. Returns 40362 error. API returns code 10003. The zse-ck anti-bot blocks all non-browser requests. Workaround: ask user to provide content manually, or search cached/mirrored versions.

Pitfalls

Images lost during extraction: WeChat wraps images in <section> > <figure> > <span> > <img>. Must recurse into sections and handle <img> before recursing into text children.
Headings missed: WeChat uses inline styles (font-weight, font-size) rather than <h1>-<h6> tags. Must check style attributes, not just tag names.
Execute_code sandbox lacks packages: bs4 (BeautifulSoup) is NOT available in the execute_code sandbox. Use terminal with conda-activated Python instead.
f-string backslash error: Python f-strings cannot contain backslashes in the expression part. Use a variable for regex patterns instead.

Reference Files

references/vocabulary-glossary-guide.md — How to create plain-language glossaries for technical articles
references/hermes-context-explanation-guide.md — How to explain abstract concepts using the user's own project as analogy