Extract useful metadata from a web article

Just how do they make those Flipboard/Twitter/Facebook previews anyway
Contents

Sample URL:

https://www.themarshallproject.org/2015/07/02/is-google-more-accurate-than-the-fbi

Sample Slack preview:

image

image

Facebook linter

https://developers.facebook.com/tools/debug/

https://developers.facebook.com/tools/debug/og/object?q=https%3A%2F%2Fwww.themarshallproject.org%2F2015%2F07%2F02%2Fis-google-more-accurate-than-the-fbi

image

image

image

What does Facebook’s scraper see?

https://developers.facebook.com/tools/debug/og/echo?q=https%3A%2F%2Fwww.themarshallproject.org%2F2015%2F07%2F02%2Fis-google-more-accurate-than-the-fbi

Raw mirror file: file