The wallabag.it bot
If you see this user agent in your logs, it’s us:
Mozilla/5.0 (compatible; wallabag; +https://wallabag.it)
Requests come from these IP addresses:
2a01:4f8:172:25b0::2
138.201.122.115
The same list in JSON: wallabag.it/bot/ips.json.
What it does
wallabag.it is a read-it-later service. When one of our users saves a link, our server fetches the page and keeps the article text, so they can read it later, on any device.
- It only fetches pages that a user asked to save. It doesn’t crawl your site and doesn’t follow links.
- A user can reload an article: the page is then fetched again.
- If a user follows one of your RSS feeds in wallabag.it, the feed is checked at most once an hour.
- The article is only shown to the user who saved it. We don’t republish it, we don’t sell it, and we don’t use it to train AI models.
Every fetch is started by a person, like opening the page in a browser. That’s why the bot doesn’t read robots.txt.
Why its own user agent
We’d rather say who we are than pretend to be a browser. If your site blocks this user agent, our users get an empty article instead of the text.
You run a website?
- To let our users save your articles, allow this user agent in your firewall or bot protection (Cloudflare, for example).
- An article is badly extracted, or you’d rather we didn’t fetch your site? Write to helpdesk@wallabag.it. We can also adjust the extraction rules for your site.
wallabag is open source: the fetching and extraction code is in wallabag and Graby.