fetch

name	fetch
description	Fetch a URL and extract its main text content as markdown.

cd ~/.claude/skills/fetch && pkgx uv run scripts/fetch.py "https://example.com"

Options

• --format <fmt> — markdown (default), txt, html, or json
• --include-links — keep hyperlinks
• --include-images — keep image references
• --with-metadata — prepend title / author / date

Uses trafilatura to strip nav/ads/boilerplate and return just the article body. For JS-rendered pages, use the browser skill instead.