User Agent Parser
Decode a User-Agent string into browser and version, engine, operating system, device type and known crawler.
Parse a User-Agent string
Prefilled with your own browser's string. Paste another to decode it; nothing is sent anywhere.
Paste a User-Agent string to decode it.
What the User Agent Parser does
User Agent Parser reads the User-Agent header a browser or bot sends and tells you what it means: the browser name and version, the rendering engine, the operating system and its version, whether the device is a desktop, phone or tablet, and whether the string belongs to a known crawler such as Googlebot, Bingbot, the Facebook link preview fetcher or GPTBot.
The field is filled with your own browser's string when the page opens, which is a quick way to see what sites learn about you. Paste any other string to decode it: one from a server log, an analytics report, a support ticket or a security alert. The parsing uses a hand-written table of patterns ordered so that Edge is not mistaken for Chrome, Chrome is not mistaken for Safari, and in-app browsers (Facebook, Instagram, WhatsApp) are named separately.
Nothing is sent anywhere; the matching runs in the page. Note that since Chrome 110 and Safari 17 the string is frozen or reduced, so the OS version and device model it reports can be a generic value even on a current device.
Step by step
- 01Read your own browser's string, which is prefilled, or paste another one into the field.
- 02The result updates as you type: browser, engine, OS, device type and bot detection.
- 03For a log analysis, paste one string at a time and compare the bot result with the IP's reverse DNS to catch fakes.
Common questions
Browsers froze or reduced the User-Agent string to limit fingerprinting. Chrome always reports "Windows NT 10.0" or "Mac OS X 10_15_7", and Android Chrome reports a generic "K" model. The true values are only available through Client Hints, which a page has to request. The parser shows what the string says and marks these known frozen values.
Anyone can send any User-Agent, and scrapers often borrow Googlebot's. The string alone proves nothing. Confirm the IP: Google publishes its crawler ranges, and a real Googlebot address has a reverse DNS name ending in googlebot.com that resolves back to the same IP. Use the Reverse DNS Lookup for that check.
Since iPadOS 13, Safari requests the desktop version of sites by default and sends a Mac User-Agent string. There is no reliable way to tell such an iPad from a Mac from the string alone; the parser reports what the string claims and notes the ambiguity.
The engine renders the page: Blink in Chrome, Edge, Opera, Brave and most others; WebKit in Safari and every browser on iOS; Gecko in Firefox. Two browsers on the same engine display pages identically, so for compatibility testing the engine matters more than the brand.
Keep checking
Reverse DNS Lookup
Find the PTR host name of an IPv4 or IPv6 address and check that it resolves back to the same address.
Free toolEmail Header Analyzer
Paste raw email headers to see the delivery path hop by hop with delays, SPF, DKIM and DMARC results and spoofing warnings.
Free toolRobots.txt Tester
Fetch a site's robots.txt and test whether a path is allowed or blocked for Googlebot or any user agent, with the matching rule.
Free toolWebsite Down Checker
Check from our server whether a website is up: status code, response time, redirects, server header and DNS.
Free toolRedirect Checker
Trace a URL's redirect chain hop by hop, spot loops, meta refresh and JS redirects, and compare the http, https and www variants.
Want these fixes done for you?
Our team runs the audit, makes the changes and reports the result.