Extract readable main content from web pages as markdown via Defuddle CLI.
npx skills add https://github.com/carlos-algms/dotfiles --skill defuddle-carlos-algms
Please help me install this Agent Skill. Skill: defuddle Source: https://github.com/carlos-algms/dotfiles/tree/main/AI-configs/skills/defuddle Command: npx skills add https://github.com/carlos-algms/dotfiles --skill defuddle-carlos-algms
Web pages often contain navigation, ads, and clutter that waste tokens and make it harder to read or analyze the content you actually care about.
Run defuddle parse <url> --md to extract the page into clean markdown.
High-intent search queries and answers about installing and using this skill.
To extract main content from web pages and convert it to markdown, run defuddle parse <url> --md. This removes navigation and clutter, producing clean text for articles, blog posts, and documentation while saving tokens.
The best way to get clean text from online documentation is using the Defuddle CLI with markdown output mode. It specifically targets readable main content, stripping away navigation and ads to reduce wasted tokens.
Yes, you can pull targeted metadata like title and description from a URL. Defuddle supports extracting specific summary attributes and domain information when you only need summary fields rather than the complete page text.
You should not use web content extraction for URLs ending in .md. Native markdown files should be processed directly to avoid unnecessary parsing and ensure the original markdown structure remains intact.
Yes, web scraping for markdown extraction works on standard web pages and blog posts. The tool supports workflows for online documentation and articles, extracting readable main text while removing clutter to optimize token usage.
vibewatchFetch URLs and extract readable text from HTML pages and PDFs.
family3253Crawl websites and save pages as local Markdown files.
georgekhananaevCompress Markdown and text into token-efficient formats with Python scripts.
OpenAEC-FoundationDeploy ERPNext/Frappe websites via REST API with Page Builder.
