defuddle-web-cleaner

Extract and clean readable article content, metadata, and markdown from URLs or HTML for research, note taking, and web scraping.

---

name: defuddle-web-cleaner description: extract clean article content from web pages using defuddle. use when a user provides a url or html and wants the readable article text, markdown version, or structured metadata. helpful for web scraping, research workflows, note taking, obsidian clipping, and converting web pages to markdown.

Extract the main readable content from a web page.

This skill removes unnecessary elements such as:

• navigation bars
• sidebars
• ads
• comments
• footers
• social buttons

The result is clean article content.

Supported Inputs

1. URL
2. Raw HTML
3. Web page text

Output Format

Default output:

Title
Author
Site
Published date

Markdown article content

Alternative output (JSON):

{ title, author, site, description, published, content, contentMarkdown }

Processing Steps

1. Detect input type
2. Load page HTML
3. Run Defuddle parser
4. Extract metadata
5. Convert to Markdown if requested
6. Return clean content

Example

Input:

https://example.com/blog/ai

Output:

Title: AI is Changing Everything
Author: Jane Smith
Site: Example Blog

Markdown:

AI is Changing Everything

Artificial intelligence is transforming industries...

Tips

Use this skill when:

• saving articles to Obsidian
• building research datasets
• cleaning webpages for LLM processing
• summarizing articles