User Agent Parser

What this tool does

User Agent Parser breaks down any User-Agent header into its browser, rendering engine, operating system, device and CPU architecture using ua-parser-js, and flags known bots, crawlers and AI assistants such as GPTBot. Parse a single string or paste thousands of lines at once to get a browser, OS and device-type breakdown with CSV export, or load your own browser’s current User-Agent.

The User-Agent header looks like a wall of version numbers and parenthesised tokens, but it encodes exactly which browser, rendering engine, operating system and device sent a request, information that log analysis, analytics and bot filtering all depend on. Reading it by eye is unreliable because browsers deliberately mention other engines’ names for compatibility, so Chrome, Edge and many others all reference Safari and Gecko.

This parser runs the ua-parser-js library locally to extract the real browser, engine, OS, device and CPU architecture, and layers in bot-detection rules that recognise known crawlers, command-line tools and, specifically, AI crawlers and assistants such as GPTBot and ClaudeBot. Single-string mode shows a full breakdown with the raw JSON; bulk mode parses up to 20,000 lines at once and summarises the most common browsers, operating systems and device types, with a CSV download of every parsed row.

How to use User Agent Parser

1. Paste a User-Agent string into the input, pick one of the Examples, or click Use my browser’s user agent.
2. Switch Mode to Bulk and paste one User-Agent per line to analyse a batch instead of a single string.
3. Read the Browser, Operating system, Device type, Device, Rendering engine and CPU architecture cards for a single string.
4. Watch for the automated-traffic notice, which distinguishes an ordinary bot or crawler from an AI crawler or assistant.
5. In bulk mode, review the summary tables for the most common browsers, operating systems and device types.
6. Download the parsed results as a CSV file for further analysis in a spreadsheet, or inspect the raw JSON for one entry.

Tips for better results

• A User-Agent naming "Safari" does not guarantee WebKit; Chrome and Firefox both include Safari in their strings for legacy reasons, which is why Browser type is still worth checking.
• AI crawler and assistant detection covers automated fetchers such as GPTBot and ChatGPT-User, not every request that merely mentions an AI product in its string.
• Mobile Chrome on Android and Chrome-based browsers on desktop report different device types, so bulk summaries are a quick way to estimate your traffic’s device mix.
• curl, wget and other command-line clients send very short User-Agent strings with no engine or OS tokens at all, which is normal rather than an error.

Common use cases

• Auditing server logs to see which browsers and operating systems real visitors use before dropping support for an old one.
• Checking whether a spike in traffic comes from a legitimate browser, a scraping bot, or an AI crawler indexing content for a model.
• Debugging a layout bug reported by a user by decoding the exact browser version and OS from their support ticket.
• Building an allow list or block list for crawlers by first seeing which bots already show up in existing logs.

Common mistakes to avoid

• Trusting the User-Agent header as a security control; it is sent by the client and can be changed to anything, so it should inform decisions, not enforce them.
• Assuming a missing device type means desktop; some desktop browsers simply omit mobile-specific tokens rather than confirming a device category.
• Blocking every string containing the word "bot" without checking the bot-detection result first, which can also catch monitoring tools you actually want to allow.