wechat-article-extractor

What This Skill Does

The wechat-article-extractor skill is a specialized tool designed to bridge the gap between WeChat's proprietary web ecosystem and standard structured data formats. WeChat articles, hosted on the mp.weixin.qq.com domain, are notoriously difficult to parse due to dynamic rendering and frequent platform updates. This skill automates the heavy lifting by programmatically fetching, parsing, and cleaning article content. It excels at extracting granular metadata such as author identity, publication timestamps, cover images, and the full HTML body of the post. Beyond simple text retrieval, it handles sophisticated link parameters like biz, mid, and idx, which are essential for tracking or cross-referencing content across the WeChat ecosystem. It effectively transforms inaccessible web content into structured JSON objects, enabling seamless integration into AI pipelines or personal knowledge management systems.

Installation

To integrate this skill into your OpenClaw environment, execute the following command in your terminal: clawhub install openclaw/skills/skills/freestylefly/wechat-article-extractor-skill Ensure that your OpenClaw agent has the necessary network permissions to reach the WeChat server endpoints, as this skill performs external requests to resolve URLs.

Use Cases

• Research Automation: Automatically aggregate industry insights from multiple WeChat Official Accounts into a centralized database or Notion page.
• Content Archiving: Convert ephemeral social media posts into local structured records for offline accessibility or long-term historical analysis.
• Content Curation: Build newsletters or recommendation feeds by pulling article titles, descriptions, and cover images without manual copy-pasting.
• Marketing Monitoring: Track specific authors or topics by parsing published content and analyzing frequency or keyword density.

Example Prompts

1. "Extract the content and author information from this WeChat link: [URL] and format it as a summary."
2. "Get the metadata for the article at [URL], specifically looking for the publish time and cover image URL."
3. "Can you parse the latest article from [URL] and tell me if it contains a repost or video content?"

Tips & Limitations

To ensure consistent results, it is recommended to provide the full canonical URL. If you encounter issues with access limits, consider using the HTML injection method by fetching the page through your own proxy first and passing the raw HTML to the extractor. Note that the skill is dependent on WeChat's frontend structure; if the platform changes its DOM structure, updates to the skill may be required. Furthermore, content marked as private or restricted by the author will return the standard error code; ensure the URL is publicly accessible before running the extraction.