User-Agent Analyzer

Instantly parse any User-Agent string to identify the browser, OS, device, and bots like Googlebot.

Common UA Examples

The User-Agent is a value that anyone can freely change and send. The results here are estimates based on string patterns, so use them for reference only. Do not use them as a basis for access control or billing. Newer Chrome-based browsers intentionally reduce UA information for privacy reasons (UA Reduction).

What is the User-Agent Analyzer?

Ever wonder what those cryptic 'Mozilla/5.0...' lines in your web logs mean? This User-Agent Analyzer decodes any UA string, instantly revealing the browser, OS, device type, and rendering engine. It's especially useful for identifying traffic from search crawlers like Googlebot, AI bots like GPTBot, or users in in-app browsers from Facebook or KakaoTalk. Because all processing happens directly in your browser, you can safely paste logs without your data ever leaving your computer. This free tool requires no installation or sign-up.

How to use

  1. Paste a User-Agent string into the "User-Agent String" input box. The analysis appears instantly as you type.
  2. To check your own browser, click the "๐Ÿ–ฅ Insert My Browser UA" button.
  3. To test a common string, browse the "Common UA Examples" and click a preset like "iPhone Safari" or "Googlebot".
  4. View the summary of the Browser, Operating System, Rendering Engine, and Device in the cards that appear.
  5. For more details, like device model, CPU architecture, or bot type, check the "Detailed Analysis" table below.

User-Agent Analyzer guide

How this tool is used in real work, and what to watch out for.

Anyone Can Forge a User-Agent

The User-Agent is just a string a browser sends to say, "This is who I am." There is absolutely no validation. It can be changed with a few clicks in developer tools, or with a single line on the command line.

bash 
# Send any UA you want โ€” there's no validation
curl -A "Mozilla/5.0 (iPhone; CPU iPhone OS 17_5 like Mac OS X)" https://example.com

# To verify a real Googlebot, you must reverse-lookup its IP
host 66.249.66.1                       # โ†’ Should resolve to crawl-...googlebot.com
host crawl-66-249-66-1.googlebot.com   # Then do a forward lookup (must match the original IP)
Do not use the User-Agent as a basis for access control, billing, or security decisions. Most scraping bots use a standard Chrome UA. Conversely, legitimate users might appear to have a strange UA due to privacy extensions. This is why all results from this tool are labeled as "estimates."

The Messy History of the User-Agent String

Why does Chrome claim to be Mozilla and Safari? It's the result of a 30-year-long game of one-upmanship, all packed into a single string.

  • Netscape started with `Mozilla/1.0`. At the time, servers used the UA to check for feature support and would only serve enhanced pages to 'Mozilla'.
  • To get those pages, IE masqueraded as `Mozilla/4.0 (compatible; MSIE 4.0)`. This started the convention of putting another browser's name first and your own identity in parentheses.
  • Safari forked KHTML to create WebKit and added "KHTML, like Gecko." This was to get pages intended for Gecko (the engine used by Firefox).
  • Chrome forked WebKit to create Blink but kept the `AppleWebKit` and `Safari` tokens to continue receiving pages made for Safari.
  • As a result, today's Chrome UA contains Mozilla, AppleWebKit, KHTML, Gecko, Chrome, and Safari. The only part that's actually true is 'Chrome'.
This is why this tool's engine detection identifies a UA as Blink if it sees `Chrome` or `Edg` tokens alongside `AppleWebKit`. If you took the string at face value, you'd mistake Chrome for a WebKit-based browser.

In-App Browsers: A Frequent Source of Failure

When you click a link in an app like KakaoTalk (a Korean messenger), Instagram, or the Naver App (a Korean portal), it opens in a built-in browser. It might look like Chrome, but it's actually a functionally limited WebView, which causes certain things to fail.

  • File downloads might not work or may fail silently. A classic example is trying to download a PDF quote or invoice.
  • New windows opened with `window.open` are often blocked. This kills payment modules and identity verification pop-ups.
  • Adding to the Home Screen (PWA installation) is impossible, and permissions for the camera or location are tied to the app's permissions, leading to unexpected behavior.
  • Cookies and local storage are sandboxed within each app, so login sessions don't persist across browsers.
This tool detects an "In-app Browser" by looking for `KAKAOTALK`, `Instagram`, `FBAV`, `NAVER(inapp)`, `Line`, `DaumApps`, `Coupang` tokens, and the Android WebView marker (`wv`). The usual solution is to display an informational banner. On Android, you can force-open Chrome using the `intent://` scheme. On iOS, there's no reliable method, so the most practical advice is to guide users to "Open in Browser" via the top-right menu.

Bot Detection: What to Block and What to Allow

If you don't filter out bots from your server logs, your visitor stats will be completely misleading. However, you can't just block every bot; they serve very different purposes.

You can paste the UA of each bot from this tool's preset chips to see how it's identified.

CategoryExamplesHow to Handle
Search CrawlersGooglebot, Yeti (Naver), Daumoa, bingbotBlocking them will get you de-indexed. Do not block.
Link PreviewsKakaoTalk-Scrap, facebookexternalhit, TwitterbotIf blocked, thumbnails won't appear when shared on KakaoTalk, Facebook, etc.
Ads & MonetizationMediapartners-Google, AdsBot-GoogleRequired for AdSense ad serving and verification.
AI TrainingGPTBot, ClaudeBot, CCBot, BytespiderBlocking is a policy decision. Control via robots.txt.
SEO AnalysisAhrefsBot, SemrushBot, MJ12botConsumes traffic with no direct benefit to you. Many sites block them.
Automation & ScriptsHeadlessChrome, python-requests, curlJudge based on its purpose.
If users complain that link previews aren't showing up when they share on KakaoTalk, first check if `KakaoTalk-Scrap` can access the page. In most cases, a `robots.txt` file, firewall, or bot-blocking rule is accidentally blocking this crawler.

User-Agent Information is Disappearing (UA Reduction)

To protect privacy (and prevent fingerprinting), Chrome is intentionally reducing the amount of information in the User-Agent string. It's no longer as detailed as it used to be.

  • The browser's minor version is frozen at `0.0.0`. In `Chrome/126.0.0.0`, the last three digits are now meaningless.
  • The Android device model is replaced with "K". This tool leaves the model value blank if it sees "K" or "wv."
  • The Android OS version is also sometimes reported as a fixed value, so you can't determine the device generation from the version alone.
  • On Mac, you can't distinguish between Intel and Apple Silicon via the UA. Even Apple Silicon Macs report "Intel Mac OS X."
If you need more accurate information, the new standard is to use Client Hints (`navigator.userAgentData`) instead of the UA string. However, it's only supported in Chromium-based browsers (not Safari or Firefox), so you should design with the assumption that neither method is perfect.

Frequently asked questions

Are the analysis results 100% accurate?

No. A User-Agent string can be easily faked. The results are estimates based on known patterns and should be used for reference only, not for critical systems. Modern browsers also reduce UA data for privacy (UA Reduction).

Is it safe to paste server logs here?

Yes, absolutely. This tool runs entirely in your web browser. Nothing you type or paste is ever sent to our servers, so your log data remains completely private to your computer. When you close the page, the data is gone.

How does this tool identify bots and crawlers?

The tool contains patterns for many known bots, including Googlebot, GPTBot, and Bingbot. When you paste a UA string, it's checked against this list. If a match is found, it's labeled as a bot and its purpose is described.

When would I need to analyze a User-Agent?

It's mainly for developers and site admins. It helps diagnose issues specific to a certain browser, analyze traffic sources from server logs, and identify whether unusual traffic is from a person, a search crawler, or a malicious bot.

More Web Tools

URL Encoder / Decoder Safely percent-encode strings for URLs or decode them back to a human-readable format. HTML Entity Converter Escape special characters like <, >, & into HTML entities, or unescape entities like &amp; back into characters. URL Parser Break down any URL into its components, edit query parameters in a table, and rebuild the new URL instantly. Markdown Live Preview Write and preview Markdown in real-time with a live editor that converts your text to HTML instantly in your browser. MIME Type Dictionary Quickly find the correct MIME type (Content-Type) for any file extension, or look up extensions by MIME type. Meta Tag Generator Generate SEO, Open Graph, and Twitter meta tags from your page's title, description, and image, with live search and social media previews.