After many unsuccessful attempts to find an XSS in Yahoo’s domains, I decided to move my attention to Microsoft Bing.
If you have a Microsoft account, Bing allows you to save online content (images, videos and places) on the page My saves, and allows to create collections to better manage your own content.
The titles of these collections were not properly filtered, so it was possible to break the code and inject persistent arbitrary code.
The code could be injected easily, all it took was the wrong image added to My saves.
I was lucky with Bing, now I can go back to fail with Yahoo 🙂
Attack Scenario
An attacker pastes on a site (for example a popular blog) the link of an image found on Bing. The GET parameter q has a malicious javascript code. Example:
www.bing.com/images/search?view=detailV2&ccid=ugtTKk%2bf&id=77C601FDB8F3CAC089D09B4F80E5984B3D66972F&thid=OIP.ugtTKk-ffUvl93ztAPX8mAHaD8&q=malicious_javascript&simid=608017275115016031&selectedIndex=0&ajaxhist=0
A victim visits this site randomly, sees the image, opens it and saves it in My Saves. Bing uses the value of variable q to give a title to the new collection. The malicious code runs whenever the victim opens My saves.
In this attack there is no interaction between the attacker and the victim.
Proof of Concept
16/04/2018 – I send the report
19/04/2018 – The vulnerability is fixed and I’m rewarded by having my name written in the Hall of Fame
21/04/2018 – Public disclosure