Category: Vulnerabilities

Dec 04 2014

Hacking PayPal Accounts With One Click

PayPal CSRF HackYasser H. Ali, an Egyptian security researcher, has discovered three critical vulnerabilities in PayPal website which could be used to take control of customer accounts.

Yasser explained in his post the three vulnerabilities used to breach into accounts:

1] Reusable CSRF Token :
The CSRF token “that authenticate every single request made by the user” which can be also found in the request body of every request with the parameter name “Auth” get changed with every request made by user for security measures, but after a deep investigation I found out that the CSRF Auth is Reusable for that specific user email address or username, this means If an attacker found any of these CSRF Tokens, He can then make actions in the behave of any logged in user.

2] Bypassing the CSRF Auth System :
an Attacker can obtain the CSRF Auth which can be valid for ALL users, by intercepting the POST request from a page that provide an Auth Token before the Logging-in process, check this page for the magical CSRF Auth “https://www.paypal.com/eg/cgi-bin/webscr?cmd=_send-money”. At this point the attacker Can CSRF “almost” any request on behave of this user.

3] ByPassing the Security Questions Change :
the request of setting up the security questions “which is initiated by the user while signing up” is not password-protected, and it can be reused to reset the security questions up without providing the password, hence, Armed with the CSRF Auth, an attacker can CSRF this process too and change the victim’s Security questions.

At this point, An attacker can conduct a targeted CSRF attack against a PayPal users and take a full control over his account Hence, An attacker can CSRF all the requests including but not limited to:

― Add/Remove/Confirm Email address
― Add fully privileged users to business account
― Change Security questions
― Change Billing/Shipping Address
― Change Payment methods
― Change user settings (Notifications/Mobile settings)…. and more.

Here is the Proof Of Concept Video:

The vulnerability is patched very fast and PayPal paid me the maximum bounty they give ($ 10,000 USD) ;) ” – Yasser Ali

Original Story: Yasser Ali’s Blog >> Hacking PayPal Accounts with one click

Dec 01 2014

WhatsApp – Message Handler Vulnerability Allows to Remotely Crash

WhatsApp Remote CrashA Vulnerability has been discovered in the wildly popular messaging app WhatsApp, which allows anyone to remotely crash WhatsApp just by sending a specially crafted message, two security researchers reported.

Two India based independent security researchers, Indrajeet Bhuyan and Saurav Kar, both 17-year old demonstrated the WhatsApp Message Handler vulnerability.

In a video demonstration, they showed that by sending a 2000 words (2kb) message in special character set can crash the receiver’s app. The worried impact of the vulnerability is that the user who received the specially crafted message will have to delete his/her whole conversation and start a fresh chat, because opening the message keeps on crashing WhatsApp unless the chat is deleted completely.

What makes it more serious is that one needs to delete entire chat with the person they are chatting to in order to get back whatsapp work in normal“, Bhuyan told.

According to the duo, the reported vulnerability has been tested and successfully works on most of the versions of Android Operating system including Jellybean, Kitkat, and all the below android versions.

Similarly, Any member of your WhatsApp group could intentionally send a specially crafted message to exit people from the group and delete the group. Also, for example, if I don’t want someone to keep records of my chat with them, then I can also send the same message exploit to the person.

The vulnerability has not been tested on iOS, but it is sure that all versions of WhatsApp including 2.11.431 and 2.11.432 are affected with this bug. Also the attack does not work on Windows 8.1.

Proof-of-Concept (PoC) Video

WhatsApp, bought by Facebook for $19 billion in February this year, has 600 Million users as of October 2014, and according to the researchers, an estimated number of users affected by the vulnerability could be 500 Million.

WhatsApp was in news recently for making end-to-end encryption on all text messages as a default feature in an effort to boost the online privacy and security of its users around the world. The app maker describe this move as the “largest deployment of end-to-end encryption ever.”

Sep 16 2014

Twitter Vulnerability Allows Hacker to Delete Credit Cards from any Account

Twitter Delete Credit Card VulnerabilityAt the beginning of this month, just like other social networks, Twitter also started paying individuals for any flaws they uncover on its service with a fee of $140 or more offered per flaw under its new Bug Bounty program, and here comes the claimant.

An Egyptian Security Researcher, Ahmed Aboul-Ela, who have been rewarded by many reputed and popular technology giants including Google, Microsoft and Apple, have discovered a critical vulnerability in Twitter’s advertising service that allowed him deleting credit cards from any Twitter account.

Actually he found two vulnerabilities not one but both was having the same effect and impact.

#First Vulnerability
The first vulnerability he had spotted was in the delete functionality of credit cards in payments method page

https://ads.twitter.com/accounts/[account id]/payment_methods

Twitter Vulnerability Delete Credit Card

When he choose to delete credit card and press on the delete button, an ajax POST request is sent to the server like the following:

POST /accounts/18ce53wqoxd/payment_methods/destroy HTTP/1.1
Host: ads.twitter.com
Connection: keep-alive
Content-Length: 29
Accept: /
Origin: https://ads.twitter.com
X-CSRF-Token: Lb6HONDceN5mGvAEUvCQNakJUspD60Odumz/trVdQfE=
X-Requested-With: XMLHttpRequest
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/37.0.2062.94 Safari/537.36
Content-Type: application/x-www-form-urlencoded; charset=UTF-8
Referer: https://ads.twitter.com/accounts/18ce53wqoxd/payment_methods
Accept-Encoding: gzip,deflate
Accept-Language: en-US,en;q=0.8
Cookie: [cookies here]
account=18ce53wqoxd&id=219643

There are only two post parameters sent in request body-
account: the twitter account id
id: the credit card id and it’s numerical without any alphabetic characters

All he had to do is to change those two parameters to his twitter account id and credit card id, then reply again the request and he suddenly found that credit card have been delete from the other twitter account without any required interaction.

The funny part that the page response was “403 forbbiden” but the credit card was actually deleted from the account :D

#Second Vulnerability
Aboul-Ela found another similar vulnerability but this time the impact was higher than the previous one.
when he tried to add an invalid credit card to his twitter account it displayed an error message

“we were unable to approve the card you entered” and show a button called “Dismiss”

Hacking Twitter Delete Credit Cards

When he pressed on the Dismiss button the credit card was disappeared from his account, so he thought it have the same effect of deleteing, so he tried to add invalid credit card again and intercepted the request which looks like the following:

POST /accounts/18ce53wqoxd/payment_methods/handle_failed/220152 HTTP/1.1
Host: ads.twitter.com
Connection: keep-alive
Content-Length: 108
Cache-Control: max-age=0
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8
Origin: https://ads.twitter.com
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/37.0.2062.94 Safari/537.36
Content-Type: application/x-www-form-urlencoded
Referer: https://ads.twitter.com/accounts/18ce53wqoxd/payment_methods
Accept-Encoding: gzip,deflate
Accept-Language: en-US,en;q=0.8
Cookie: [Cookies Here]
utf8=%E2%9C%93&authenticity_token=Lb6HONDceN5mGvAEUvCQNakJUspD60Odumz%2FtrVdQfE%3D
&id=220152&dismiss=Dismiss

This time account parameter doesn’t exists and only credit card id is used.

So he changed the id in the url and body to his credit card id from other twitter account then replied the request, and guess what?
Credit card got deleted from the other twitter account !

VIDEO DEMONSTRATION:

Dec 15 2013

EBay Remote Code Execution Vulnerability Demonstrated

EBay Exploit
A German Security researcher has demonstrated a critical vulnerability on Ebay website.

He found a controller which was prone to remote-code-execution due to a type-cast issue in combination with complex curly syntax.

In a demo video, he exploited this RCE flaw on EBay website, and managed to display output of phpinfo() PHP function on the web page, just by modifying the URL and injecting code in that.

According to an explanation on his blog, he noticed a legitimate URL on EBay:

https://sea.ebay.com/search/?q=david&catidd=1

and modified the URL to pass any array values including a payload:

https://sea.ebay.com/search/?q[0]=david&q[1]=sec{${phpinfo()}}&catidd=1

Video Demonstration:

David has already reported the flaw responsibly to the Ebay Security Team and they have patched it early this week.

Source: eBay : Remote Code Execution

Oct 01 2013

New Touch ID Hack Allows to Unlock iPhone by Multiple FingerPrints

iPhone 5S Touch ID HackThe Iranian group defeated the very basic phenomenon of an iPhone Fingerprinting scanner, which allows them to unlock an iPhone device with multiple Fingerprints.

Apple‘s iPhone 5s, was launched just available in stores two weeks before with a new feature of biometrics-based security system called “Touch ID”, that involves analyzing a user’s fingerprint and using that to unlock the phone.

Apple launched the technology that it promises will better protect devices from criminals and snoopers seeking access. With this you can purchase things from the iTunes App Store. Basically, you can now use it in place of your password.

“Fingerprint is one of the best passcodes in the world. It’s always with you, and no two are exactly alike,” according to the Apple’s website.

Last week Germany Hackers showed that how they were able to deceive Apple’s latest security feature into believing they’re someone they’re not, using a well-honed technique for creating a latex copy of someone’s fingerprint.

Another interesting fact is that, Touch ID is not only designed to scan the fingerprints of your fingers, it works with various human body parts and appendages which are also not fingers.

In a video demonstration, the Group set up a mixed Fingerprint scan of 5-6 people for an iPhone 5S handset (as shown in the video), which allowed all of them to unlock the locked device with their individual fingerprint.

According to Apple, the chance that Touch ID will misread a finger is 1 in 50,000 , this is because Touch ID is not designed to capture the fingerprint in strict mode. It scans the fingerprint on a very high-resolution (2400 dpi), to get and match the partial parts of an impression for faster unlocking.

If the iPhone is not able to scan the thumb impression in the strict mode to be unique, there is a possibility that out of 1000 thumb impressions iPhone’s Touch ID system can count 2-3 impressions as of the same person.