Web Application Security

Chapter 11. Cross-Site Request Forgery (CSRF)

Sometimes we already know an API endpoint exists that would allow us to perform an operation we wish to perform, but we do not have access to that endpoint because it requires privileged access (e.g., an admin account).

In this chapter, we will be building Cross-Site Request Forgery (CSRF) exploits that result in an admin or privileged account performing an operation on our behalf rather than using a JavaScript code snippet.

CSRF attacks take advantage of the way browsers operate and the trust relationship between a website and the browser. By finding API calls that rely on this relationship to ensure security—but yield too much trust to the browser—we can craft links and forms that with a little bit of effort can cause a user to make requests on his or her own behalf—unknown to the user generating the request.

Oftentimes CSRF attacks will go unnoticed by the user that is being attacked—as requests in the browser occur behind the scenes. This means that this type of attack can be used to take advantage of a privileged user and perform operations against a server without the user ever knowing. It is one of the most stealthy attacks and has caused havoc throughout the web since its inception in the early 2000s.

Query Parameter Tampering

Let’s consider the most basic form of CSRF attack—parameter tampering via a hyperlink.

Most forms of hyperlink on the web correspond with HTTP GET requests. The most common of which is simply an <a href="https://my-site.com"></a> ...