Network & DNS
User Agent Parser
Parse a browser user-agent string into browser, version and operating system.
Results appear here.
What the result means
The box starts with this browser's navigator.userAgent. You can replace it and parse any other string. The page splits nothing onto a server. Tokens are matched with ordered rules: Edge before Chrome, Opera before Chrome, then Chrome, Chromium, Firefox, and Safari. The first matching rule wins, which is why order matters. Operating system rules look for Windows, Android, iOS, macOS, ChromeOS, and Linux. Device type is a guess from words such as Mobile, iPad, or bot. Strings that look like bots, and strings that match none of the browser rules, stay Unknown. That is the honest result, not a failure of the page. The raw string is shown beside the guess so you can see what the rules used. Versions are taken from the token the rule names, and odd clients will not be forced into a brand.
Related tools
Questions
What is a user-agent?
It is a header browsers and other clients send to describe themselves. Sites use it to guess a browser and an operating system. It is not a password and it is easy to change.
Why is the device a guess?
The string is not a structured form. Rules look for familiar tokens. A tablet can say Mobile, and a desktop browser can be set to impersonate a phone. The label on the result says the guess is heuristic.
Why is a bot marked unknown?
Crawlers use many shapes, and forcing them into a browser name would be a guess. If the string looks like a bot, or matches no rule, the browser stays Unknown.