Fetch web content as Markdown using a four-tier fallback strategy.
npx skills add https://github.com/Airmomo/skills --skill smart-web-fetch
Please help me install this Agent Skill. Skill: smart-web-fetch Source: https://github.com/Airmomo/skills/tree/main/smart-web-fetch Command: npx skills add https://github.com/Airmomo/skills --skill smart-web-fetch
This Skill eliminates the manual effort of obtaining web content in a clean, Markdown-friendly format, reducing noise from raw HTML and saving tokens during AI processing.
Fetch the article at https://example.com and return the cleaned Markdown content.
High-intent search queries and answers about installing and using this skill.
You fetch web content as clean Markdown by replacing raw HTML retrieval with a dedicated cleaning layer. This Skill orchestrates a four-tier fallback to reliably return readable Markdown, minimizing token usage for agents and automation tasks.
The best way to reduce token usage is to clean web content into Markdown before processing. This Skill removes HTML noise and returns structured Markdown or JSON, ensuring agents only consume the essential article text for summarization or analysis.
The fallback strategy works by sequentially trying Jina Reader, markdown.new, and defuddle.md before reverting to original HTML. This maximizes content availability by automatically cascading to the next cleaning provider if the previous one fails.
You need access to the Jina service or equivalent clean-content providers for the primary fallback tier. If Jina is unavailable, the Skill automatically falls back to markdown.new, defuddle.md, and finally the original HTML to ensure content retrieval.
Yes, you can get structured JSON output instead of Markdown. The Skill processes fetched web pages through its cleaning layer and can produce either clean Markdown text or structured JSON, depending on your agent or script requirements.
NigarumOvumConvert web page content from URLs into clean Markdown text.
smallnestConvert PDFs, Word, Excel, HTML, and images with OCR into Markdown.
lookleeCompress or trim chat history to minimize token consumption.
JKHeadleyFetch clean readable web content from URLs via llms.txt and markdown fallback.
