User Agent Parser

How to use it

1. Copy a User-Agent string from server logs, devtools or an analytics export.
2. Paste it into the input box — parsing updates as you type.
3. Read the five result fields: browser, engine, operating system, device type and whether it is a crawler.
4. Crawler names such as Googlebot and bingbot are recognised from a built-in list.
5. Fields the UA does not reveal show "unknown" — the tool reports what is there, nothing more.

Common use cases

• Reading access logs — identifying which browser and OS produced a request.
• Debugging compatibility reports — checking what an old or unusual UA actually declares.
• Writing bot filters — seeing how common crawler User-Agents identify themselves.
• Understanding UA claims vs reality — noting that a UA is a self-declared string, not proof.

Why it stays private

Matching is done with regular expressions inside your browser. The string is never sent anywhere.

FAQ

Q: Can it be wrong? A: Yes. A User-Agent is a self-reported string that browsers can spoof and old conventions muddle — Windows NT 10.0 covers both Windows 10 and 11, and Opera copies Chrome's UA. The parser marks low-confidence fields as unknown rather than inventing an answer, but treat results as the UA's claim, not ground truth.

Q: Which browsers are recognised? A: Chrome, Edge, Firefox, Safari, Opera and Internet Explorer, with version numbers when the UA carries them. Engines map to Blink, Gecko, WebKit and Trident. Other or future browsers show as unknown.

Q: Does it detect every crawler? A: No. It checks a built-in list of about twenty well-known crawlers plus a generic /bot|crawl|spider/ pattern. Custom or unnamed scrapers will show as a regular client.