VorzaBot

VorzaBot is the crawler behind Vorza, the Agent Experience platform. It reads public pages to measure how retrievable and citable a product is to AI coding agents: the site itself, robots.txt, llms.txt, documentation pages, package registry metadata, and public GitHub pages. It always identifies itself with the user agent above and never impersonates another crawler.

Fetches happen only for brands a Vorza customer has registered in their own workspace. VorzaBot follows no links beyond the pages it checks, submits no forms, executes no code, and sends no credentials.

Limits it obeys

• robots.txt is obeyed: a path your rules disallow for VorzaBot (or for all crawlers) is not fetched, and the affected measurements report not measured with the robots reason.
• At most one request per second to any one domain, strictly serialized.
• At most 40 fetches per brand per crawl and 300 per crawl, with an 8 second timeout and a 2 MB response cap.
• Crawls run at most a handful of times per week per workspace, and a brand with fresh results is not re-fetched inside the freshness window.

Opting out

Disallow VorzaBot in your robots.txt and it will not fetch the disallowed paths:

User-agent: VorzaBot
Disallow: /

Affected checks then report not measured with the robots reason rather than guessing. Questions: admin@vorza.dev.