tjdwls101010

ultra-fetch

Fetch, crawl, or map web pages into clean, context-efficient markdown saved to a file — using a stealth browser that reaches sites the built-in WebFetch can't (bot-protected, Cloudflare, JS-rendered), plus BM25 filtering that keeps only the parts relevant to a query. Use this as the default for actually reading a web page's content, following a search result deeper, reading across a whole site, or discovering what URLs a site has — especially after a WebSearch, or whenever WebFetch is blocked, returns junk, or you need the result saved to disk. NOT for a trivial quick fact where WebFetch already suffices, NOT for logged-in or authenticated pages (out of scope — use the dedicated scrape-x / scrape-fb tools for X and Facebook), and NOT for developing the ultra-fetch tool itself, which is ordinary repo work.