User Agent Parser

How to use

1. Paste a user agent string — from a log file, an analytics report, or a support ticket.
2. Or press Use mine to read your own browser's string.
3. Read the browser, engine, operating system and device it claims to be.
4. Copy any single part, or the whole string.

Features

• Correctly separates Edge, Opera, Samsung Internet and Brave from the Chrome they all claim to be.
• Takes Safari's real version from Version/, not the WebKit build number.
• Translates Windows NT 10.0 into Windows 10 or 11, and iOS 17_6_1 into 17.6.1.
• Tells tablets from phones, which the strings make deliberately awkward.
• Runs in your browser — log lines you paste are not uploaded.

Frequently asked questions

Why does every browser claim to be Mozilla?

Because of thirty years of compatibility workarounds. Sites once checked for "Mozilla" before serving modern HTML, so every browser added it; then sites checked for Safari, so Chrome added that; then for Chrome, so Edge added that. A user agent string is a pile of claims about what a browser is compatible with, not a statement of what it is. This parser tests the most specific impostor first, which is the only way to read them.

Can I trust what it says?

No, and neither can a website. A user agent is sent by the client and can say anything — browsers offer to change it, and crawlers routinely copy a browser's. Treat it as a claim, useful for statistics and support, not as identification.

Is this bot detection?

No. It reports strings that announce themselves as crawlers — Googlebot, curl, python-requests. Anything pretending to be a browser is indistinguishable here, which is exactly why this reports and does not block.

Why is my Windows version shown as "10 or 11"?

Because Windows 11 reports itself as Windows NT 10.0, the same as Windows 10. Microsoft froze the value deliberately, so the string genuinely cannot tell them apart — and a tool that picked one would be guessing.

Related tools