# Onto-Reader · About our fetcher > Onto-Reader/1.0 is an on-demand fetcher, not a crawler. What it is, what it does, and the exact robots.txt to allow or block it. **Source:** /about-bot **Extracted:** 2026-08-28T19:19:40.235Z --- Onto-Reader ## An on-demand fetcher, _not a crawler._ You found this because our user-agent string points here. It is one GET, made only when a customer asks for that exact URL, and it reads your robots.txt first. Everything you need to allow or block it is on this page. The string in your logs `Onto-Reader/1.0 (+https://buildonto.dev/about-bot)` Origin network Vercel edge · api.buildonto.dev Operator Onto — buildonto.dev Contact founder@buildonto.dev 1 Request per call 15s Hard timeout 1 hr Cache before refetch Behaviour ### What it does, and what it never does. The second column is the part most people actually want to know. #### It does * Fetches a URL only when an authenticated customer requests that exact URL * Reads /robots.txt before every fetch and honours a block * One GET per call, 15-second hard timeout, follows redirects * Caches for one hour, so a repeated request doesn't hit you twice * Surfaces 403 / 401 / timeout to the caller as a named error #### It never * Spider, follow links, or build an index * Execute JavaScript, accept cookies, or submit forms * Queue background jobs against your origin * Retry silently after your server refuses it Control it ### Two lines to stop it. One caveat to know. Onto-Reader re-reads robots.txt on the next request, so a change takes effect immediately and there is nothing to notify us about. #### Allow it everywhere The default. Nothing to add — Onto-Reader treats an absent rule as permission, the same as any well-behaved agent. User-agent: \* Allow: / #### Block it entirely Onto-Reader matches on GPTBot. A global disallow stops every fetch, and the API returns the policy state to whoever asked. User-agent: GPTBot Disallow: / #### Path rules won't stop it Worth knowing: Onto-Reader currently checks for a site-wide disallow only. Per-path rules like these are read but not enforced against the fetch, so a route you want kept out needs a site-wide block or real authentication. User-agent: GPTBot Disallow: /admin/ Disallow: /account/ Allow: / Why it exists ### It reads for one person at a time. Onto turns a page into clean Markdown so an agent can use it without downloading a megabyte of markup. That only happens when someone asks for your URL specifically — there is no index being built, and nothing is stored beyond the one-hour cache. If Onto-Reader is in your logs, one of our customers pointed an agent at your page. Blocking it blocks that read; it does not affect any other crawler, and it does not remove you from anything. If it behaved in a way this page doesn't describe, that is a bug and we want to hear about it — [founder@buildonto.dev](mailto:founder@buildonto.dev). While you're here ### See what an agent gets from your site. The same extraction Onto-Reader performs, run on your URL, with a score for how readable it came back. [Scan your site](/scanner) [How the fetch works](/how-it-works)