We have a ASP.NET MVC application. All the POST requests (form submits) have been protected from CSRF by using @Html.AntiForgeryToken
and ValidateAntiForgery
In short, what you've just described is not an example of an XSRF attack...
Both CSRF
and XSRF
are used to describe what's called a Cross Site Request Forgery
. It's where a malicious website takes advantage of your authenticated state on another website, to perform fraudulent cross-site requests.
Imagine that you're authenticated
on your bank's website, and that your banks website contains a form
to create new transactions, all pretty straight forward...
The Malicious website
Now let's think of the Malicious website
you're also visiting, imagine that it also contains a form
, one that is hidden and the values of which are pre-populated...
When the form on the malicious website is submitted, an HTTP request
will be sent straight from you to your bank, and because you're authenticated on your bank's website, the transaction could be accepted.
Essentially, an attacker is using your own authentication against you by forging requests and using you as the messenger to deliver that request.
You use an anti-forgery token, this token
is a string containing a random value, the token is placed in your cookies
, in addition to your HTML forms.
When you receive a request, you validate that the form contains an anti-forgery token and that it matches the one stored in your cookies. A malicious site can not see the tokens your website sets on a client, and without this information, XSRF attacks are stopped in their tracks.
On your controller Action that will be handling the request, add the attribute [ValidateAntiForgeryToken]
, and in the HTML form add (@Html.AntiForgeryToken())
.
public class ExampleController : Controller
{
[ValidateAntiForgeryToken]
[HttpPost]
public ActionResult Test(Foo fooModel)
{
// do your thing...
return this.View();
}
}
That's it!
Anti-Forgery Tokens don't make a lot of sense when performing GET
requests, in fact, they don't make sense to have them anywhere that you're not modifying and persisting data, as any GET request will be returned to your user, not the attacker.
If you're Creating, Updating or Deleting data... make sure that you're using it then.