Verso's crawler

If you found Verso/1.0 (+https://verso.so/bot) in your logs, that was us fetching a feed. Verso is a feed reader operated by Fair Copy L.L.C. in Seattle, and this page is here so you can find out who we are, and reach a person, from one line of your access log.

Why we fetched you

Because someone subscribed to your feed. Verso only requests URLs its readers have added to their accounts — it does not spider links, it does not discover sites on its own, and it does not follow its way across your domain. If we're polling you, a person asked us to.

We are not an AI crawler. Nothing Verso fetches is used to train machine-learning models, or sold, or handed to anyone else. It is parsed into entries and shown to the person who subscribed.

How often we ask

At most once an hour per feed, and usually less. Every request is conditional — we send back the ETag and Last-Modified you gave us, exactly as you gave them, so an unchanged feed costs you a 304 and no body. Polls are jittered by up to ten minutes so a pile of feeds imported together doesn't stay in lockstep and hit you on the same tick.

If your feed says how often it wants to be polled — RSS <ttl>, the syndication module, skipHours, skipDays — that beats our own schedule. We never poll a feed more often than it asks for. Accounts still on a free trial are slowed further in proportion to how many feeds they hold.

We also open only one connection to a given host at a time. If one reader imports forty feeds from your server, they are fetched one after another, not all at once.

How we handle a "no"

429 and 503: we read Retry-After and honor it — as seconds or as a date — for up to a week. It isn't treated as a suggestion.

Errors: each consecutive failure doubles the wait, up to a day between attempts. A feed that stays broken is checked daily, not hourly.

410 Gone: we stop. Permanently, without further requests.

301 and 308: we adopt the new address, so the redirect is paid for once rather than on every poll.

What else we request

Alongside the feed, Verso may fetch your site icon, and one page at a time on a reader's direct action: the full text of an entry your feed only summarized, or a page they saved to read later. Both are a single request for a single address, made because somebody pressed a button, and neither leads anywhere else on your site. Images inside articles are fetched by us and re-served from our domain, which is why those requests come from Verso's address rather than a reader's: your server never sees who is reading you. That's a privacy decision on our side, not an attempt to hide.

Making us stop

The fastest way is to block or rate-limit the Verso/ User-Agent at your server; we treat a 403 as an error and back off from it.

Or email support@verso.so and a person will answer: to have us slow down, to have us drop a feed, or to tell us we're misbehaving. If Verso is costing you real traffic, that's a bug on our side and we want to know about it.