User Agent Parser
Parse a User-Agent string into browser, version, rendering engine, operating system and device type (desktop, mobile, tablet or bot). Paste one UA per line to parse several at once, with a JSON export.
Input
Output
Parsed result
| Browser | Browser Version | Engine | OS | OS Version | Device Type |
|---|---|---|---|---|---|
| No data yet | |||||
JSON export
Guides
The User Agent Parser breaks a raw User-Agent (UA) string down into the things you actually want to know: which browser and version made the request, its rendering engine, the operating system and version, and the device type — desktop, mobile, tablet, or a bot/crawler.
Every HTTP request a browser sends carries a User-Agent header, a dense one-line string like Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36. It is packed with historical tokens ("Mozilla", "AppleWebKit", "KHTML, like Gecko") that make it hard to read at a glance. This tool applies an ordered set of pattern-matching rules to pull out the meaningful fields and present them clearly.
How to use it
- Paste a User-Agent string into the input box.
- Read the parsed Browser, Browser Version, Engine, OS, OS Version and Device Type in the results table.
- Grab the structured JSON export if you need the data in code, or download the table as CSV.
To analyze many UA strings at once, paste one per line — each line becomes its own row in the table, and the JSON export becomes an array of objects. This is handy for combing through server access logs.
What it detects
- Browsers: Chrome, Firefox, Safari, Edge, Opera, Samsung Internet, Vivaldi, Yandex, UC Browser, Internet Explorer and more, including their iOS variants (which all run on WebKit).
- Engines: Blink, WebKit, Gecko, Trident, EdgeHTML and Presto.
- Operating systems: Windows (mapped from the NT kernel version, e.g.
NT 10.0→ Windows 10), macOS, iOS, Android, Chrome OS, Linux and common distributions. - Device type: desktop, mobile, tablet, or bot.
- Bots & crawlers: search engines (Googlebot, Bingbot, DuckDuckBot, Baidu, Yandex), AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Amazonbot), SEO tools (SemrushBot, AhrefsBot) and social scrapers (Facebook, Twitter, Slack, Discord).
Why does the version sometimes look wrong or say "—"?
A User-Agent string is self-reported by the client and can be spoofed, frozen, or deliberately vague. A dash (—) simply means that field is not present in the string — many bot and command-line UAs, for example, carry no OS or engine token at all. Windows 11 is a known blind spot: it reports the same NT 10.0 token as Windows 10, so it is indistinguishable from the UA alone.
Is my data sent anywhere?
No. Parsing runs entirely in your browser — the User-Agent strings you paste never leave your device.
Where do I find a User-Agent string?
Your own browser's UA is visible in its developer tools (the Network tab shows request headers), and server access logs record the UA of every visitor. This tool intentionally parses whatever string you paste rather than auto-reading your current browser, so you can inspect any UA — including one copied from a log or a bug report.