defuddle

Name: defuddle
Availability: InStock
Author: gabrielcardosodeaguiar45-oss

Extract clean markdown from web pages using the Defuddle CLI.

1|Updated May 10, 2026

Search Tags:#markdown #web scraping #content extraction #token optimization #defuddle #online reading

Ready to run

npx skills add https://github.com/gabrielcardosodeaguiar45-oss/claude-skills-azevedolima --skill defuddle-gabrielcardosodeaguiar45-oss

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: defuddle
Source: https://github.com/gabrielcardosodeaguiar45-oss/claude-skills-azevedolima/tree/main/defuddle
Command: npx skills add https://github.com/gabrielcardosodeaguiar45-oss/claude-skills-azevedolima --skill defuddle-gabrielcardosodeaguiar45-oss

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Web pages often contain navigation, ads, and clutter that waste tokens and make summaries harder, so you need clean, readable content.

Core Features & Use Cases

Clean Markdown Extraction: Uses Defuddle CLI to strip away clutter and return readable markdown.
Smaller, More Useful Context: Reduces token usage by focusing on main content instead of page chrome.
Targeted Metadata: Extracts specific fields like title, description, or domain when you only need key attributes.

Quick Start

Run Defuddle on the page you want to analyze using the command: defuddle parse <url> --md

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 web pages for AI context?▼

You can extract clean markdown from web pages by running `defuddle parse <url> --md`. This strips navigation, ads, and page clutter to reduce token usage and return readable content for AI analysis.

Why does web scraping for AI waste so many tokens?▼

Web scraping for AI wastes tokens because standard HTML pages contain navigation, ads, and page chrome clutter. Defuddle solves this by extracting only the main readable markdown content, significantly reducing token consumption for AI context windows.

Can I extract specific metadata like title and description from HTML pages?▼

Yes, you can extract specific metadata like title, description, or domain from HTML pages. Defuddle parses web pages to return targeted fields instead of full content when you only need key attributes for online documentation.

Does web content extraction work for online documentation and blog posts?▼

Yes, web content extraction works for online documentation, articles, blog posts, and standard HTML pages. Defuddle processes these page types to remove clutter and deliver clean markdown for AI reading and analysis.

When should I avoid using markdown extraction tools for URLs?▼

You should avoid using markdown extraction tools for URLs ending in `.md` files. Defuddle skips web scraping for these direct markdown-file URLs and retrieves the raw markdown content directly.