$ docker run -ti -p 127.0.0.1:5000:5000 blabla1337/owasp-skf-lab:url-redirection-harder
Now that the app is running let's go hacking!
Running the app Python3
First, make sure python3 and pip are installed on your host machine. After installation, we go to the folder of the lab we want to practise "i.e /skf-labs/XSS/, /skf-labs/jwt-secret/ " and run the following commands:
$ pip3 install -r requirements.txt
$ python3 <labname>
Now that the app is running let's go hacking!
Docker Image and write-up thanks to ContraHack!
Reconnaissance
Step 1
The application shows that there is a new version of the website available somewhere, and a click on the button "Go to new website" will redirect you to it.
Intercepting the traffic generated by the application, we note that the redirection is performed using the following call
GET /redirect?newurl=newsite
that will generate a 302 Redirect response from the server
Exactly like in the previous example (KBID XXX). If we look at the code we discover a tiny difference: a blacklist!
landing_page = request.args.get('newurl')
if blacklist(landing_page):
return render_template("index.html", content ="Sorry, you cannot use \".\" in the redirect")
return redirect(landing_page,302)
If we look at the blacklist definition, we can immediately see that the URL, in order to be valid, must not contain any "." (dot).
defblacklist(url):
blacklist =["."]
for b in blacklist:
if url.find(b)!=-1:
returnTrue
​
returnFalse
Step 3
Let's verify the effectiveness of this blacklist. If we try to exploit the unvalidated redirect using an external website, we see that the application blocks us, returning an error in the page.
If we URL encode the dot the application is smart enough to decode it and recognise it in the URL, blocking us again.
Exploitation
Although we cannot explicitly use the dot character, we can find different ways to bypass the blacklist. In example we could use the following techniques:
double encoding: https://google%252ecom
UTF-8 encoding: https://google.com%E3%80%82.com
Can you find more?
Using the payload above we will be able to successfully redirect a user to a malicious website