This site is the archived OWASP Foundation Wiki and is no longer accepting Account Requests.
To view the new OWASP Foundation website, please visit https://owasp.org

Testing for Cross site scripting

From OWASP
Revision as of 15:19, 12 October 2006 by Mmeucci (talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

OWASP Testing Guide v2 Table of Contents

This article needs to be written or reviewed.
Please contact Matteo Meucci if you are interested to contribute.

This article is a stub. You can help OWASP by expanding it or discussing it on its Talk page.



Cross Site Scripting (XSS) vulnerabilities

Cross site scripting is an attack on a client (yes you)which uses vulnerablities in websites a user may visit. It is commonly used in Phishing and identity theft attacks. The key to a vulnerable cross-site script is such that: (in a nutshell): The web site or web application redisplays text inputted by the user without any/proper data validation. When script (javascript normally) is entered as a parameter value it is redisplayed to the user or to anyone that activates a link to redisplay the inputted data.

How to Test

Black Box testing

Example: A search function on a web application: When the "Search" button is pressed a HTTP request is sent:

GET http://www.weaksite.com/page.jsp?id=09586&searchparam=my%20favorite%20colour

This request is a typical GET request with a search parameter "searchparam". searchparam contains the search criteria/payload for the search.

The result of the search is returned to the user, redisplaying the search criteria entered: The HTML returned by the server after performing a search (acting on the http request includes:

<h1>Could not find: </h1>my%20favorite%20colour

You can see that the user input passed via the searchparam query string parameter was probably placed in a string variable on the server and inserted by the Web application into a customised <h1> tag.

If the application is not validating input server-side (forget client-side validation, thats not security!!), an attacker could abuse this in many ways:

One can make the searchparam value to be interprited at part of the page markup (ie Javascript) rather than it being simply a value inputted by the user.

Lets try:

GET http://www.weaksite.com/page.jsp?id=09568&searchparam=</h1><textarea></textarea>

Here searchparam terminates the <h1> tag with </h1> this effectivley breaks out of a line of HTML. Then the addition on the string <textarea></textarea> is actually interprited as a piece of HTML markup and a text area field shall be displayed.

Taking this a step further:

again, breaking out of the <h1> tag:

GET http://www.weaksite.com/page.jsp?id=10&lang=en&searchparam=</h1><script>alert(‘Hello’)</script>

The HTML output from this last request would look like:

<h1>Could not find: </h1><script>alert(‘Hello’)</script>

The browser would interpret the java script as part of the page as opposed to user input and execute and alert box with the word 'Hello' in it.

<h1>Could not find: </h1><script>alert(‘Hello’)</script>


OWASP Testing Guide v2 Table of Contents