Amazon has awarded an $18,000 bug bounty for an exploit chain that could have allowed an attacker to take complete control of a Kindle e-reader simply by knowing the targeted user’s email address.
The attack, dubbed KindleDrip, was discovered in October 2020 by Yogev Bar-On, a researcher at Israel-based cybersecurity consulting firm Realmode Labs. KindleDrip involved the exploitation of three different security holes, all of which were addressed by Amazon.
The first vulnerability in the exploit chain was related to the “Send to Kindle” feature, which allows users to send an e-book in MOBI format to their Kindle device via email as an attachment. Amazon generates an @kindle.com email address where a user can send e-books as an attachment from a list of email addresses approved by the user.
Bar-On discovered that he could abuse this feature to send a specially crafted e-book that would allow him to execute arbitrary code on the targeted device. The malicious e-book achieved code execution by leveraging a vulnerability related to a library used by the Kindle to parse JPEG XR images. Exploitation required the user to click on a link inside an e-book that contained a malicious JPEG XR image, which would result in a web browser opening and the attacker’s code getting executed with limited privileges.
The researcher also discovered a vulnerability that allowed him to escalate privileges and execute the code as root, which gave him complete access to the device.
“The attacker could access device credentials and make purchases on the Kindle store using the victim’s credit card. Attackers could sell an e-book on the store and transfer money to their account,” Bar-On explained in a blog post. “At least the confirmation email would make the victim aware of the purchase.”
It’s worth noting that an attacker could not gain access to actual credit card numbers or passwords through such an attack because this type of data is not stored on the device. Instead, they could obtain special tokens that can be used to access the victim’s account.
An attacker would have only needed the targeted user’s email address and to convince the victim to click on a link inside the malicious e-book. While the Send to Kindle feature only allows users to send e-books from pre-approved email addresses, the researcher pointed out that an attacker could have simply used an email spoofing service. The prefix of the @kindle.com email address of the targeted user is in many cases the same as their regular email.
The security holes that required changes to the Kindle firmware — the code execution and privilege escalation issues — were patched in December with the release of version 5.13.4. Amazon now also sends a verification link to email addresses that cannot be authenticated, it adds a few characters to some email aliases to make them more difficult to guess, and systems are in place to prevent brute-forcing of the email address. Kindle users do not need to take any action.
A video has been published to show how a KindleDrip attack worked:
“The security of our devices and services is a top priority. We have already released an automatic software update over the Internet fixing this issue for all Amazon Kindle models released after 2014,” an Amazon spokesperson told SecurityWeek. “Other impacted Kindle models will also receive this fix. We also have measures in place to help prevent customers from receiving content they haven’t requested. We appreciate the work of independent researchers who help bring potential issues to our attention.”
*Updated with statement and clarifications from Amazon