Researcher demonstrates Twitter XSS vulnerability

A Twitter user has demonstrated a cross-site scripting (XSS) vulnerability on the microblogging platform that could allow an attacker to take over users' accounts or spread malware.

An Indonesian security researcher, using the alias “H4x0r-x0x” and Twitter handle “0wn3d_5ys,” discovered the vulnerability and demonstrated the bug using his own Twitter account.

In addition, the researcher on Monday announced details about the flaw on a blog.

The vulnerability affects the “application name” field on Twitter's application registration page, used by developers when setting up a new Twitter application.

The flaw appears to be the result of a lack of input validation of the “application name field” when accepting new requests for Twitter applications, Daniel Kennedy, partner at Praetorian Security Group, told SCMagazineUS.com on Thursday. The flaw could be exploited by cybercriminals to insert malicious JavaScript code into a Twitter page.

“I haven't seen it used by attackers yet, but obviously that can change,” Kennedy said.

Visiting the researcher's Twitter account causes a pair of XSS alert boxes, followed by a user's browser being manipulated. The demonstration of the flaw also causes an animation from the film “The Matrix” to appear, followed by messages from the researcher, one of which states, “My Twitter Owned By : H4x0r-x0x..”

“Infection [account takeover] can be accomplished simply by visiting a profile with an include of a malicious JavaScript, making a true self propagating website worm possible,” according to a post on Praetorian Prefect.

A Twitter spokesperson told SCMagazineUS.com on Thursday that the company is aware of the issue and has fixed it for new applications, but is still working to patch it in all programs.

Last August, a separate but similar XSS bug affecting Twitter was discovered by software developer James Slater. In that case, Twitter's application programming interface (API), used by developers to create applications to post tweets, did not properly filter the URL of these programs. As a result, users could actually insert malicious JavaScript code, along with a URL.


See original article on scmagazineus.com