Clone to Claude
Last commit on March 3, 2026·Created on March 3, 2026
joeseesun/defuddle-skill
“Pulls the article out from under the ads.”
Combined rank
#550
across all skills
In Automation
#35
category rank
Stars
107
0.0% in last 7d
Forks
10
0.0% in last 7d
Watchers
0
0 watchers in last 7d
Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of real attention, adoption, and active maintenance.
Description
Defuddle extracts the main text from a web page or local HTML file, stripping away ads, sidebars, navigation, and everything that gets in the way of reading. It returns clean Markdown with metadata including title, author, source, and word count. The skill is built for researchers, journalists, and anyone who needs to clip, archive, or reuse article content without fighting a page's layout.
WHO IT'S FOR
Content Researchers
extract clean article content for analysis
Journalists and Writers
clip and save web articles with metadata
AI Agent Developers
integrate web content extraction into AI workflows
Knowledge Management Teams
curate and archive web articles into knowledge bases
30 Days of GitHub Stars
+1 stars↑ 0.9% monthly change
Jul 7Jul 14Jul 21Jul 28Aug 5
Compatibility & Install1 Skill · 1 Compatible agent · 3 categories · 3 install paths
Compatible agents
❯Claude Code
Categories
AutomationWritingData
Install viaRepository instructions Repository instructions
Repository instructions
Manual Copy
NPX Skills Add
Also in this category4 of 34