This is the SeeItSaveIt add-on.

See ianb.github.io/seeitsaveit for a more detailed end-user explanation.

This addon helps extraction scripts. The scrapers act on what you are looking at, not just the URL you are looking at. The difference? If you extract data from a URL you can only scrape the public and non-personalized web, what another server will see when they ask for the same page you are looking at.

The addon works by waiting for you to hit the extract button, then it asks a server (as implemented trivially in server/) for any scrapers that know how to scrape that URL. It asks you which scraper you want to use. The whole page is then frozen and passed on to the scraping code. This is not like Greasemonkey or a bookmarklet, in that the scraper never touches the live page you are looking at.

The scraper can then produce some JSON of whatever it might want to scrape from the page. Then you are asked where you want to send that JSON - scraping and the consumption of the scraped content are separated.

I roughly maintain a list of ideas for enhancing this project in the TODO file.