Skip to main content

Getting started with Contextractor

The fastest way to try Contextractor is one command — no install, no API key, and no browser download:

npx contextractor extract-one https://example.com/ --crawler-type cheerio

npx runs the published contextractor package without installing it. extract-one prints one page's cleaned content straight to stdout, so you can pipe it anywhere. --crawler-type cheerio fetches the page over plain HTTP and parses it with Cheerio — no headless browser is downloaded or launched, which is what makes this the lightest way to start. You only need Node 22+.

The trade-off is that Cheerio sees only the server-rendered HTML. For JavaScript-heavy pages that need a real browser, install Chromium once with npx playwright install chromium, then drop --crawler-type cheerio — the default adaptive crawler switches between plain HTTP and the browser per page.

Build a more advanced command in the playground

When you want more — a whole-site crawl, specific output formats, proxy rotation, or a crawl-now/export-later storage pipeline — the playground is the quickest way there. Set the options visually, choose npm CLI (or Apify, or the library) under "What to do", and it generates the exact ready-to-run command or code for your configuration. Copy it and run it locally.

Next steps

  • npm CLI — the full command-line reference: extract, extract-one, export, purge, flags, and JSON config.
  • npm library — call extractOne(url) or createExtractor() from your own Node.js code.
  • Apify Actor — run Contextractor on Apify at scale, with no local setup.

Updated: July 5, 2026