ZetselBot
Zetsel reads a fixed list of feeds each morning to make one edition. If you have found ZetselBot in your logs, this is what it was doing and how to stop it.
What it reads
Three things, and nothing else. The feed you publish, once an hour at most and usually far less, using conditional requests so an unchanged feed costs you a 304. The page behind an item in that feed, once, to read its text. And the head of that page, once, to find the image you already set as your Open Graph image.
It does not crawl. It never follows a link out of your article, never visits a page that was not in your feed, and keeps no copy of your article beyond a short extract used to decide whether the piece is worth a place in the paper. Every story in every edition links back to you, under your own headline.
How it identifies itself
ZetselBot/1.0 (+https://www.zetsel.app/bot)
Mozilla/5.0 (compatible; ZetselBot/1.0; +https://www.zetsel.app/bot)
The second is browser-shaped in the convention Googlebot uses, because some servers reject anything that does not look like one. The identity inside it is honest either way.
How to keep it out
Add this to your own robots.txt and it will stop, on its next pass, without being asked twice:
User-agent: ZetselBot
Disallow: /
That line is the whole opt-out for now: Zetsel has not been announced yet, and a contact address is published with the announcement. Anything disallowed is dropped on the next pass, no reason required.
Who runs it
Kraftwrk, Rotterdam. Zetsel is a daily paper for product designers, made by one person, and the source list is chosen by hand. If your feed is in it, that was a deliberate compliment.