BrillixBot

Last updated: October 15, 2026

BrillixBot is the web crawler that Brillix uses to build its own search index. It reads pages that are open to the public, so that Brillix can find them when someone asks a question and link back to them as a source.

How to identify it

Every request from BrillixBot carries this user agent:

The product token is BrillixBot. A request that uses this name but does not behave as described below is not ours, and we would like to hear about it.

What it does

• It fetches public web pages and feeds and reads their text.
• It keeps a headline, a short summary and the start of the page text, so the page can be found by search and linked to.
• It follows links from pages it is allowed to read, up to a limited depth.
• It does not sign in, fill in forms or read pages that need a password.

How it behaves

• robots.txt. It reads a site's robots.txt before fetching from it and follows the rules written for BrillixBot. When rules for both a specific path and a wider path match, the longest rule wins, and an allow wins a tie.
• Crawl-delay. It leaves at least one second between requests to the same site. If your robots.txt asks for a longer Crawl-delay it waits that long, up to 30 seconds.
• Daily budget. Each site has a limit on how many pages are fetched per day, and BrillixBot stops when the limit is reached.
• Size and time. It reads at most 3 MB from a page and gives up on a request after 15 seconds.
• Public addresses. It refuses to connect to private or internal network addresses, even when a name points at one.
• Page-level controls. It respects noindex and nofollow in a robots meta tag or a brillixbot meta tag, and it does not follow links marked rel="nofollow", ugc or sponsored.

How to block it

To keep BrillixBot out of an entire site, add this to https://your-site/robots.txt:

To keep it out of part of a site, list those paths under Disallow instead of /. To keep a single page out of search, add <meta name="brillixbot" content="noindex"> to it. Changes are picked up the next time BrillixBot reads your robots.txt, which is at least once a day while it visits.

Contact

Questions, complaints or removal requests: bot@brillix.ai. Please include the address of the page or the site and, if you have them, a few lines from your server log. We answer every message.

For how Brillix handles personal data more generally, see the Privacy Notice.