AntSpaceBot

Version 2026-10-18, in force since October 18, 2026

AntSpaceBot is AntSpace's web reader. It reads a company's public website to draft or update that company's profile on AntSpace: what it offers, where it is, its clients and events.

How to recognise it

  • Every request it makes, a page, a sitemap, a document or a picture, says Mozilla/5.0 (compatible; AntSpaceBot/1.0 (+https://antspace.ai/bot)).
  • Pages that only show their content with JavaScript are read in a headless Chrome browser that sends the same user agent and runs the page's scripts as a visitor's browser would. It doesn't load the page's images, video or fonts.
  • It runs from Amazon Web Services addresses in the United States (Ohio).

When it visits

  • Only when there's a reason to read a particular site: someone asks AntSpace to add a company, a company's own team runs autofill on its profile, or we start a profile for a company from public information.
  • It reads a few pages of a site (the home page and up to about ten others, chosen from its navigation and its sitemaps: products, services, pricing, about, customers and contact pages), some documents those pages link to, and the company's logo and pictures. It doesn't read the whole site, and it doesn't come back on a schedule.
  • It reads one page of a site at a time, waits between requests (at least a second, longer when your robots.txt asks for it with Crawl-delay), and backs off when a site answers 429 or 503, as long as Retry-After asks.
  • It never signs in, fills in forms, solves CAPTCHAs or other human checks, or gets around paywalls. It skips sign-in, account, cart and checkout pages. A page that answers with a block or a human check is left out, never used as what the company says.

robots.txt

AntSpaceBot follows robots.txt (RFC 9309). It obeys the group for AntSpaceBot when there is one, otherwise the group for every crawler (User-agent: \*), and its Crawl-delay (up to two minutes). To keep it out of your whole site:

User-agent: AntSpaceBot
Disallow: /

It also honours text and data mining reservations (the tdm-reservation header or meta tag, /.well-known/tdmrep.json, and noai): from such pages it takes only plain facts, such as the company's name, address and what it offers, and none of their text or pictures.

How to ask it not to read your site

Add the rule above to your robots.txt: it takes effect on its next visit. Or email support@antspace.ai from an address at your domain and tell us which site: we add it to the list of sites AntSpaceBot never reads.

To have a profile AntSpace started for your company corrected or removed, use "Request removal or correction" on it, or email privacy@antspace.ai; a removal also adds your site to that list. See the Pre-seeded profiles notice.

Contact

Questions about AntSpaceBot, or a problem with how it reads your site: support@antspace.ai. Please include your domain and, if you can, the time of the visit.