Headless mode and screenshots

Render before mirroring

site-scraper https://www.example.com --headless

Add --screenshot to save a full-page PNG for every page discovered during that headless crawl:

site-scraper https://www.example.com --headless --screenshot

Screenshots without a crawl

The screenshot subcommand captures one URL and does not mirror HTML or assets:

site-scraper screenshot https://www.example.com

For a batch, provide a UTF-8 text file with one URL per line. Blank lines and lines beginning with # are ignored.

# public pages
https://www.example.com/
https://www.example.com/about/

site-scraper screenshot --file urls.txt --output shots/ --concurrency 2

A real capture

This is casoon.dev as the screenshot subcommand saved it on 26 October 2026, with the default desktop viewport. The run took 9 seconds and wrote a PNG of 1440 × 5348 pixels; the image below is scaled to 960 pixels wide and converted to WebP for this page.

site-scraper screenshot https://casoon.dev/
# screenshots/casoon.dev--index.png

Use --device tablet or --device mobile for the other viewports; the CLI reference lists their sizes.

Every URL is validated before Chrome starts. Files use readable, deterministic names such as example.com--index.png and example.com--search--q-astro.png; colliding names receive a stable short hash. Existing unrelated screenshots in the output directory are kept.