llms.txt validator

Checks the site you're on against the llmstxt.org v2 spec: the required H1, the file's structure, and the two v2 link relations — `rel="describedby"` and `rel="alternate" type="text/markdown"` — read straight from the page's own `<link>` elements and, where present, its `Link:` header. That's the one thing a website-based checker can't do: reach a staging host or an intranet address that isn't reachable from the open internet at all. This extension runs in your browser, so it reaches those the same way your browser does, and it reads the v2 link relations from the page your browser already rendered, including a page you're signed in to. Fetching the llms.txt file itself is a separate step, and it does not carry your session: the file needs to be reachable without a login for the extension to see it. That's deliberate. The extension never sends your cookies or credentials anywhere, not even to the site being checked. It has one job: checking the llms.txt for the tab you're on. It does not follow the links inside that file. Checking every link would need permission for every site those links point to, and this extension does not ask for that. Link checking lives on the companion tool at serhiichuk.dev/tools/llms-txt/, and the panel links there when a file has links worth checking. The panel also writes a paste-ready prompt for a coding agent, naming what's wrong and why, so fixing a finding doesn't mean re-reading the spec first. No data collection. The extension doesn't send anything anywhere on its own: no telemetry, no analytics, no accounts. The only outbound requests it makes are to the site already open in the tab, to fetch its own llms.txt and the few files the spec relates to it. Full accounting at serhiichuk.dev/privacy/llms-txt-extension, which is this extension's privacy policy and the URL to give the Chrome Web Store submission form.