defuddle

Extract clean markdown from web pages using the Defuddle CLI.

1|Updated Apr 4, 2026

Search Tags:#markdown #cli #content #tokens #extraction #web #defuddle

Ready to run

npx skills add https://github.com/Jylhis/claude-marketplace --skill defuddle-jylhis

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: defuddle
Source: https://github.com/Jylhis/claude-marketplace/tree/main/plugins/obsidian/skills/defuddle
Command: npx skills add https://github.com/Jylhis/claude-marketplace --skill defuddle-jylhis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Many web pages contain navigation menus, ads, and other clutter that increase token usage when processed by language models. This Skill cleans the page and returns pure markdown, making downstream analysis cheaper and clearer.

Core Features & Use Cases

Clean Extraction: Uses Defuddle CLI to strip away unnecessary HTML elements and output markdown.
Metadata Retrieval: Can pull specific properties such as title, description, or domain.
Use Case: When a user needs to read an online article or documentation without extra noise, the skill provides a tidy markdown version ready for AI consumption.

Quick Start

Ask the defuddle skill to fetch and clean the article at https://example.com/tutorial.

Frequently Asked Questions about defuddle

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I extract clean markdown from a web page URL?▼

You can extract clean markdown from a URL by using a CLI tool with a --md flag to strip away navigation, ads, and HTML clutter, returning pure markdown content ready for AI consumption and reducing token usage.

What is the best way to reduce token usage when processing web content?▼

The best way to reduce token usage when processing web content is to extract clean markdown, which removes unnecessary HTML elements like ads and navigation menus before the text is processed by language models.

Do I need to install anything to extract markdown from URLs?▼

Yes, you need the Defuddle CLI installed globally via npm to parse URLs and extract markdown content, which requires a Node.js environment to run the command line tool.

Can I retrieve specific metadata like title or description from a web page?▼

Yes, you can retrieve specific metadata properties such as title, description, or domain from a web page alongside extracting the clean markdown content from the URL.

Does extracting markdown from web pages work on standard HTML articles?▼

Yes, extracting markdown from web pages works on standard HTML articles and documentation, removing noise to provide a tidy markdown version for downstream analysis.