Hashloop's web reader
If you saw Hashloop/1.0 (+https://hashloop.ai/bot) in your logs, it was someone's loop reading one public page they asked about.
What it is
Hashloop runs loops: things people want done, like renewing a passport or watching a train fare. When a loop needs a fact from a public page (a fee, a processing time, a price), it reads that one page and keeps only the facts it was asked for. Never the page itself.
It reads one page at a time, when someone's loop asks. It doesn't crawl, doesn't follow links across your site, and doesn't sign in anywhere.
What it won't do
- It honours robots.txt. It reads the rules for
Hashloopif you've written some, and the rules for everyone if you haven't. It keeps a copy for a day. - It never gets past a bot check. A challenge, a captcha, a block or a "too many requests" ends the read. It isn't retried, and it never comes back under another name.
- It reads only public sites on a short list, such as GOV.UK, the NHS, rail and parcel sites, high street shops, Which?, MoneySavingExpert, the BBC and Wikipedia. A redirect off that list ends the read.
Stopping it
Add this to your site's robots.txt. It applies within a day.
User-agent: Hashloop
Disallow: /To keep it out of part of your site, disallow just that path instead.
Contact
If it's doing something you didn't expect, write to us at hello@hashloop.ai and include a line from your logs.