PeakGEOBot
PeakGEOBot is the page reader used by PeakGEO brand diagnostics. If you found it in your access logs, this page explains who it is, what it came for, and how to turn it away.
The User-Agent it sends
Mozilla/5.0 (compatible; PeakGEOBot/1.0; +https://www.peakgeo.ai/bot)What it reads
Only the official sites of brands and their competitors: usually the home page plus pages such as About or Products. A handful of pages at a time — never a full-site crawl.
How often it reads
Only when a customer starts a diagnosis. There is no periodic sweep; a single run usually means fewer than ten requests to one domain.
What it keeps
Title, description, canonical link, site name and icon, and a short text excerpt — enough to confirm the URL really belongs to that brand. No copy of the page is stored.
What it never does
GET requests only, HTML only, at most 1 MB per page. It does not log in, submit forms, run page scripts, or collect personal data.
How to block it
The PeakGEOBot User-Agent is stable, so a rule at your CDN, WAF, or server is enough. With Nginx:
if ($http_user_agent ~* "PeakGEOBot") {
return 403;
}PeakGEOBot does not read robots.txt today, so a robots rule will not stop it — use the rule above, or just email us.
Contact us
Questions about what it read, a request to slow it down, or a site you want left alone — email us with the domain and scope and we will exclude it:
[email protected]