User-Agent parser & generator
Paste a User-Agent string to see the browser, engine, OS and device it claims — or copy a current, real UA string for testing.
Parse a string
Parsed with pattern matching, the same approach server libraries use. A User-Agent is self-reported and trivial to fake, so treat the result as what the client claims, not proof.
Current common User-Agents
Representative strings for testing and QA. Real browsers update their version numbers frequently and are moving toward frozen, less detailed User-Agents plus Client Hints — do not hard-code these anywhere long-lived.
Reading a User-Agent string
A User-Agent is a single header the browser sends with every request, historically packed with backward-compatibility tokens — which is why almost every modern UA still starts with "Mozilla/5.0" and mentions "like Gecko" or "AppleWebKit" regardless of the actual browser. The useful parts are the browser name and version, the rendering engine (Blink, Gecko, WebKit), the operating system and version, and the device tokens that identify a phone or tablet. This tool pulls those out with the same pattern-matching approach server-side libraries use, and flags known crawler and bot signatures.
It is a claim, not an identity
Anyone can set any User-Agent — antidetect browsers, scrapers and testing tools do it routinely, and every browser has a devtools switch for it. So a UA that says "Googlebot" is not proof the request came from Google (check the reverse DNS for that), and blocking by User-Agent alone stops only the laziest bots. Browsers are also freezing the UA string and shifting detail to User-Agent Client Hints, so parsing will return less over time. Use this for QA, log analysis and understanding what a client reports — not as a security control.