Extract clean Markdown from web pages using Defuddle CLI.
npx skills add https://github.com/zhengfran/dotconfig --skill defuddle-zhengfran
Please help me install this Agent Skill. Skill: defuddle Source: https://github.com/zhengfran/dotconfig/tree/main/tools/ai/agents/skills/defuddle Command: npx skills add https://github.com/zhengfran/dotconfig --skill defuddle-zhengfran
It removes distracting clutter from web pages so you can extract the useful content without wasting tokens on navigation, ads, and irrelevant layout.
Use the defuddle skill to parse a web page into Markdown by running: defuddle parse <url> --md
High-intent search queries and answers about installing and using this skill.
To extract clean text from a web page, you parse the URL to remove clutter like navigation and ads, returning only readable Markdown content. This lowers token usage for downstream analysis.
Converting online documentation into Markdown involves parsing the live web URL to isolate the main article text. This process removes non-essential layout elements, yielding a clean Markdown file ready for reading or processing.
You can reduce token usage by extracting only the core readable content from web pages while discarding navigation, ads, and irrelevant layout. This ensures the text you process is strictly necessary for your task.
Yes, you can extract targeted metadata fields like title, description, or domain from a URL. This allows you to pull only the key attributes instead of fetching the full page content when specific data is needed.
No, for URLs ending in .md, you should not use standard web page parsing. Instead, delegate the fetching of markdown-URL inputs directly to WebFetch to avoid unnecessary processing.
otrebuFetch multiple URLs with curl and aggregate content into a timestamped Markdown file.
mindmorassConverts meeting transcripts into structured summaries highlighting decisions, actions, and topics.
chuan-khunaExtract clean Markdown content from web pages using Defuddle CLI.
flaz78Fetch a web page URL and return its cleaned main text content.
