Date: Sun, 4 Nov 2007 22:03:12 +0000
From: "pdp (architect)" <pdp.gnucitizen@...glemail.com>
To: reepex <reepex@...il.com>
Cc: full-disclosure@...ts.grok.org.uk
Subject: Re: on xss and its technical merit

comments inlined! I have to cuz you inlined yours

On Nov 4, 2007 9:04 PM, reepex <reepex@...il.com> wrote:
> On Nov 4, 2007 2:41 PM, pdp (architect) <pdp.gnucitizen@...glemail.com>
> wrote:
>
> >
> > 1) XSS isnt techincal no matter how its used
> >
> >
> > Also, as buffer overflows and other attacks, which are more or less
> > related to them, attackers need to take into consideration the
> > execution flow and as such make the attack stealthier.
>
> I agree with this on a very high level but not in actual application. Having
> limited chars in a xss isnt really comparable to having limited characters
> in a buffer overflow.  having A-Za-z0-9 in xss only limits what scripting
> elements you can use while the same for bin exploiting makes you rely only
> on opcodes and addresses in that range. Writing alpanumeric shellcode
> compared to writing limited xss ( esp with the ease you can redirect to
> other pages and thus not be limited at all ) is not even a close comparison
> technically.
>
> Also "controlling execution flow" of a browser which you only control
> javascript or similar is no where near as challenging as having to control
> the execution of a binary or even moreso a kernel after you have destroyed
> much of its data and have to repair it to a usable state after.
>
>

I agree, it is more complicated but don't you think that you have most
of the tools already built for you? for example, I needed to write my
own shell like interface for firefox just to get some of these nifty
BASH tricks working when doing Web based attacks, including finding
and exploiting of XSS.

The only reason bin exploits are harder is because you have to deal
with opcodes. So, this does not mean that you are smarter... it just
means that you are nerdier. It does require a lot of effort to get
going... I agree. And I have a great respect for everyone that does
it. But I don't think that it is something I cannot personally get my
head on if I really want to. It is all about dedication, something
that I and a lot of XSS people already showed that have it in some
solid forms.

But if you are saying that JavaScript is easier to read then opcodes,
you are right!

>
> >
> > 2) people who use xss on pentests/real hacking/anything but phishing
> >
> >
> > XSS is bar far the only way to run untrusted code within the origins of a
> trusted domain
> > without having a browser vulnerability on first place. SQL Injection
> > and file inclusion attacks still exists, I deal with them on a daily
> > basis, but the attack surface is largely mitigated by various types of
> > frameworks which power most of the modern applications. However, why
> > do you need SQL Injection when you can perform the needed action on
> > behalf of the user by using XSS? It is safer and a lot stealthier. If
> > you want to change someones details or want to get some data out, XSS
> > is completely valid type of attack.
>
> With software (bin) vulns you arent only relying on a user or browser or
> anything. you have vulnerabilities in the server software or perimeter
> devices so you are cutting out any "user interaction" ( which is a very
> important thing ), but maybe i am caring too much about your wording of "bar
> far the only".
>

Bin vulns are finer and there is no doubt about that. But you have to
think creatively. You are banging on the front door which is gardded
by god knows what. How is that for a stealth? If you are spreading a
worm, ok you have no problem with that but in case you want to
penetrate a network you better think twice. First of all, you may
fail. Second, you may loose all your hard work for nothing. You are
giving away your well researched exploit. We have the tools the catch
the little beast.

It is different when it comes to XSS. XSS attacks can be tangled into
the Web so deep that you won't be able to find them unless you have
some sort of control over the remote servers, which you probably
don't. It is indirect, which means that you have to think several
steps in advance, because the vector may take any form and place. Most
of the tools are located on the Web. The data is on the Web, ok the
Intranet, when it comes to corporate stuff, but it is still based on
Web technologies.

I am not sure if you agree with me but I always say that you have to
pick the best tools for the job. So here is a question for you: If
most of the data is based on Web technologies what tools would you use
in order to get it? Buffer overflows? Common on, do you have any idea
how relevant these vulnerabilities are when it comes to the Web. They
represent in total 0.01%. On the other hand XSS represent 99% .. which
one would you pick?

>
> also with xss you are limited to the tasks that web application can do
> unlike full control of the server which allows you to do whatever you want
> and allows for much deeper penetration into the network.
>
>

I agree but most of the time attackers are after the data not a
control over the server. This so 1984.

>
>
> > the people I've seen who use XSS today, have a vast background on
> > traditional attack techniques. though, their number is very small
> > mainly because the topic hasn't reached the level of maturity as other
> > topics already have.
>
> We must know different people because the people i know that tout xss are
> people that found out about xss and sql injection and have never moved on
> and consider themselves 'security professionals'
>

well I have to tell you something. people get into a state of mind
professional psychiatrist call "comfort zone". if you know about
something and you've spent so much researching and working on it, you
will never let it go. it is as simple as that. thanks god I read
different literature apart form tech books.

>
> > Not true. If you don't know, XSS is a top priority today. It is
> > present on almost all websites/application. I am not sure who you are
> > working for and whether you are doing any pentesting but I can tell
> > you something: people are interested in XSS and they are afraid of it.
> > I must say that there is a huge gap of knowledge and understandings
> > that needs to be filled but the situation is getting better with every
> > single day. Today, companies are interested in Web2.0. They are
> > interested of the impact this technology will have on their
> > organization. There are numerous of things corporate people worry
> > about when it comes to it. XSS is one of them.
> >
>
>  ok and this is a technical debate not about people getting ripped off which
> is what businesses care about.  just because xss affects businesses alot
> does not make it anymore technical or worthwhile to 'research'
>

As I said, it can get as technical as anything else. Should we start
witht Firefox peculiarities and IE ECMA standards bugs. If you don't
know about the internals of the browser, you won't be able to get to
the interesting stuff. There are many many different kinds of XSS. We
have simple reflective XSS. Dom based XSS. The persistent kinds of
XSS. Then we have XSS in websites which result into execution of
chrome code due to shared trust. We have server based XSS as well as
client based XSS. We have cached XSS. Local XSS. Remote XSS. ETC, ETc,
Etc, etc. All of them, very different... very unique.

>
> >
> > I used to rate XSS as low sometimes as medium risk two years ago.
> > Today, if they are unauthenticated, I rate them as HIGH. Why? Open
> > your eyes. XSS is not only about getting the victim running some code.
> > There are a number of things you can do. Do you know that if CNN has
> > XSS on their site and I manage to inject some google adds and kind of
> > spread around the vector on a couple of bookmarking sites, I can make
> > tones of money. Think about it.
> >
> >  a) CNN is a very important site.
> >  b) Add Clicks will cost more.
> >  c) Social bookmarking is a way of life (look at DIGG)
> >  d) Social bookmarking sites can be spammed (research OnlyWire)
> >
> > You have all the components of a successful attack. What about forging
> > stories? Or performing Black PR? Or maybe even Black SEO? The limit is
> > only your imagination. Unfortunately, some people lack the imagination
> > so others have to show them the way.
>
> Everything you listed is related (loosely) to phishing, scamming,fraud, etc
> not to anything technical or groundbreaking.  While things like hijacking
> adsense may be interesting ( which they are ), they do not require technical
> feats to accomplish. its simple techniques which any script kiddie can
> accomplish.
>

absolutely, but imagination is part of the hacking process - something
that script kiddies lack. By having only technical skills you are
nothing more but a very powerful input device. The imagination and
creativity makes you a hacker. XSS is all about imagination plus
technical stuff as well. Finding XSS, ok most of the time simple (not
very simple for the interesting kinds). Exploiting XSS... well, you
have to know about the following:

XML, XHTML, CSS, JavaScript, ECMA Script, ActionScript, XSLT, SLT,
RDF, OWA, SWF, WSF, XPath, XQuery, XForms (where needed),  HTTP (let's
not forget about this one), DOM, Rendering Engines, XPCOM (cross-breed
XSS), SOAP, WSDL (Yes XSS is possible though services as well, think
indirect!). MathML (the esoteric kinds), RSS, ATOM, Track Backs, Ping
Backs, SVG... many many more technologies. Do you really believe that
this is simple? Do you know what? We can do a tech-quiz pub night on
these technologies. If XSS was that simple all of us should know about
them, right? The sad truth is that 80% of sec guys don't use RSS, or
simple don't know the difference between RSS and ATOM.

>
>
> >
> >
> >
> > 5) publishing xss shows your weakness and that you dont have the
> >
> >
> > publishing XSS makes you look stupid as well publishing a DoS cuz you
> > haven't investigated enough to see whether and how your findings can
> > be exploited.
>
> we agree!!
>
>
> > reepex, I am sorry but all your statements are groundless. I was
> > expecting something more from you, especially after we exchanged a few
> > private emails. sometimes, I get the feeling that you actually know
> > what you are talking about. you definitely know a few things but
> > c'mon, really... give me something juicy...
> >
>
> Yea after reading my original thing i admit it was pretty weak. i hope i
> fixed it up here.
>

you are speeding up but I want more ... so far, it has been like a
walk in the park.

-- 
pdp (architect) | petko d. petkov
http://www.gnucitizen.org

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/

Powered by blists - more mailing lists