Is markdown the recommended format for Claude?

Yes, using the --md flag is highly recommended as it provides the most readable and token-efficient format for Claude to process web information.

What is the main benefit of using Defuddle over WebFetch?

Defuddle removes navigation, ads, and clutter, providing cleaner markdown that uses fewer tokens and improves Claude's focus on actual content.

How do I install the underlying tool for this skill?

You can install the Defuddle CLI globally via npm using the command 'npm install -g defuddle' to enable this skill's functionality.

Does Defuddle support JSON output?

Yes, by using the --json flag, you can receive a structured response containing both the HTML and markdown versions of the content.

Can I extract specific metadata like page titles?

Yes, you can use the -p flag followed by the property name, such as title, description, or domain, to extract specific metadata.

Defuddle Content Extractor

Name: Defuddle Content Extractor
Author: 812lcl

by812lcl

0•

ウェブスクレイピングとデータ収集

Extracts clean, clutter-free markdown from web pages to optimize token usage and improve AI context readability.

Defuddle is a specialized content extraction skill that transforms messy web pages into clean, readable markdown. By stripping away navigation menus, advertisements, and irrelevant boilerplate, it allows Claude to focus exclusively on the core content of articles, documentation, or blog posts. This process significantly reduces token consumption and provides a higher-quality context for analysis, making it an essential tool for developers who frequently reference online documentation or research web-based resources within their workflow.

主な機能

01Metadata extraction for specific properties like title and description

02Token usage optimization by filtering out non-essential HTML

03Markdown output generation for optimal AI context integration

04Support for multiple output formats including JSON and HTML

050 GitHub stars

06Clutter removal strips ads, navigation, and menus automatically

ユースケース

01Extracting clean data for research and competitive analysis tasks

02Summarizing blog posts and articles while conserving token limits

03Analyzing online technical documentation without site-wide navigation noise

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add 812lcl/dotfiles defuddle

For use in Claude.ai and ChatGPT