User-Agent String Parser

What is a User-Agent string?

The User-Agent (UA) is a small free-form text header sent by every HTTP client telling the server what software is making the request. Browsers, search-engine crawlers, REST clients and webhooks all include one. The string is historically baroque — it begins with Mozilla/5.0 for compatibility reasons, includes a list of "compatible with…" claims, and ends with the actual product token. Parsing UA strings reliably is harder than it looks because vendors invent new tokens constantly and the format has no formal grammar; this tool decodes the string for you and surfaces the browser, operating system, device type and model, layout engine, and CPU architecture in a clean grid. UA strings are slowly being replaced by the Client Hints API on Chromium browsers, but the legacy header is still everywhere and will be for years — analytics, anti-fraud, accessibility, A/B-test bucketing, and feature switches all depend on it. Use this parser when writing tests that need a realistic UA, when debugging access logs, when verifying a crawler's identity, or simply when you want to know what a particular User-Agent string actually represents.

How to use the parser

Paste any User-Agent into the box, or click Use my UA to drop yours in. The result splits into browser, operating system, device type and model, layout engine and CPU architecture. The presets cover desktop browsers, mobile browsers, search-engine bots, social-media link previewers and dev tools — handy when writing tests or replicating a real-world request.

Preset User-Agents (one click each)

These are real (or canonical) UA strings published by each vendor. Click any preset to load it into the input and read the parse result.

Desktop browsers

Mobile browsers

Crawlers and bots

Dev tools and clients

Legacy

UA string anatomy

Token	Meaning
`Mozilla/5.0`	Compatibility prefix kept for legacy reasons; almost all browsers send it.
`Windows NT 10.0`	Windows version: 10.0 maps to Windows 10/11.
`AppleWebKit`	WebKit-based browsers (Safari, Chrome, Edge) include this engine token.
`Gecko`	Mozilla’s rendering engine, used by Firefox.
`Trident`	Internet Explorer’s legacy engine — anything below 11 is unsupported.
`Mobile`	Indicates a mobile-formatted UA, often paired with a device model.
`compatible`	Self-declared compatibility group, common in older bots and IE.

Frequently asked questions

Should I trust the User-Agent?

No. The string is set by the client and any value can be sent. Use it for analytics and feature heuristics, never for authentication or authorisation.

What replaces the UA going forward?

Client Hints (Sec-CH-UA, Sec-CH-UA-Platform, etc.) are structured headers Chromium freezes into a more privacy-respecting subset of the UA. Chrome plans to reduce the UA further over time.

Why does my browser report it’s Mozilla?

Historical accident from the late 1990s — vendors copied each other’s prefixes to avoid being blocked by sniffing scripts. Every modern browser still starts with Mozilla/5.0.

How do I detect Googlebot vs the user?

Match for the Googlebot token in the UA, then verify the request IP via reverse DNS. Anyone can send a Googlebot UA — only the IP confirms it.

Are bot UAs canonical?

Mostly. Major search engines and social-media previewers publish their UA. Smaller crawlers spoof popular browser strings, which is why backend rate-limits and IP analysis matter.

Does the parser work offline?

Yes. The parser runs locally on your device after the page first loads — your UA strings never leave the browser, are not stored, indexed, logged, or shared, and the page works even if you go offline.

User-Agent String Parser