Is there a good reference guide for deciphering http user-agent strings?

Question

We're getting some odd errors reported on our website, and are trying to find out some details on who is doing the looking. I'm looking at a sample user-agent strings and seeing things like:

Mozilla/5.0 (Windows; U; Windows NT6.0; en-US; rv:1.9.0.6) Gecko/2009011913 Firefox/3.0.6

Now some of this is obvious, but some of it is less so ("U"?) From the pattern of the hits, I'd suspect that this was a robot, but I don't see anything that I recognize in this string to clue me in to that. I'd like to know what some of the things in this (and other different) strings mean. Is there a comprehensive reference somewhere?

Answer 1

This Wikipedia page gives a good overview, and describes the meaning of the 'U' attribute as well.

Answer 2

User-agent strings (as all request headers) are not to be trusted at all, but I recommend www.useragentstring.com as a reference for helping you determine the patterns you're interested in.

Answer 3

The U stands for security as the Mozilla User Agent String Reference states. The link to wikipedia has been provided by somebody else already

In future I suggest you use these two sites

http://www.useragentstring.com/

http://www.browserscope.org/

Answer 4

Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.0.11) Gecko/2009060214 Firefox/3.0.11

This is my current user agent.
Mozilla/5.0 - browser
Macintosh - platform
U - 256 bit encryption supported
Intel Mac OS X 10.5 - my operating system
en-US - language
rv:1.9.0.11 - revision
Gecko/2009060214 - rendering engine
Firefox/3.0.11 - browser version

Answer 5

Robots can fake User-Agent strings, so what it declares is irrelevant. It can simply be done with e.g. curl -A user-agent-here.