How Malicious Android Apps Slip Into Disguise – Krebs on Safety

Researchers say cell malware purveyors have been abusing a bug within the Google Android platform that lets them sneak malicious code into cell apps and evade safety scanning instruments. Google says it has up to date its app malware detection mechanisms in response to the brand new analysis.

At subject is a cell malware obfuscation technique recognized by researchers at ThreatFabric, a safety agency based mostly in Amsterdam. Aleksandr Eremin, a senior malware analyst on the firm, instructed KrebsOnSecurity they not too long ago encountered plenty of cell banking trojans abusing a bug current in all Android OS variations that includes corrupting parts of an app in order that its new evil bits can be ignored as invalid by well-liked cell safety scanning instruments, whereas the app as an entire will get accepted as legitimate by Android OS and efficiently put in.

“There may be malware that’s patching the .apk file [the app installation file], in order that the platform remains to be treating it as legitimate and runs all of the malicious actions it’s designed to do, whereas on the identical time quite a lot of instruments designed to unpack and decompile these apps fail to course of the code,” Eremin defined.

Eremin mentioned ThreatFabric has seen this malware obfuscation technique used a number of instances prior to now, however in April 2023 it began discovering many extra variants of recognized cell malware households leveraging it for stealth. The corporate has since attributed this enhance to a semi-automated malware-as-a-service providing within the cybercrime underground that may obfuscate or “crypt” malicious cell apps for a charge.

Eremin mentioned Google flagged their preliminary Could 9, 2023 report as “excessive” severity. Extra not too long ago, Google awarded them a $5,000 bug bounty, regardless that it didn’t technically classify their discovering as a safety vulnerability.

“This was a singular scenario through which the reported subject was not categorized as a vulnerability and didn’t affect the Android Open Supply Mission (AOSP), however did lead to an replace to our malware detection mechanisms for apps which may attempt to abuse this subject,” Google mentioned in a written assertion.

Google additionally acknowledged that among the instruments it makes obtainable to builders — together with APK Analyzer — at the moment fail to parse such malicious functions and deal with them as invalid, whereas nonetheless permitting them to be put in on person units.

“We’re investigating attainable fixes for developer instruments and plan to replace our documentation accordingly,” Google’s assertion continued.

Picture: ThreatFabric.

In line with ThreatFabric, there are a number of telltale indicators that app analyzers can search for which will point out a malicious app is abusing the weak point to masquerade as benign. For starters, they discovered that apps modified on this means have Android Manifest files that include newer timestamps than the remainder of the recordsdata within the software program package deal.

Extra critically, the Manifest file itself can be modified in order that the variety of “strings” — plain textual content within the code, reminiscent of feedback — specified as current within the app does match the precise variety of strings within the software program.

One of many cell malware households recognized to be abusing this obfuscation technique has been dubbed Anatsa, which is a complicated Android-based banking trojan that sometimes is disguised as a innocent software for managing recordsdata. Final month, ThreatFabric detailed how the crooks behind Anatsa will buy older, deserted file managing apps, or create their very own and let the apps construct up a substantial person base earlier than updating them with malicious parts.

ThreatFabric says Anatsa poses as PDF viewers and different file managing functions as a result of a lot of these apps have already got superior permissions to take away or modify different recordsdata on the host machine. The corporate estimates the folks behind Anatsa have delivered greater than 30,000 installations of their banking trojan by way of ongoing Google Play Retailer malware campaigns.

Google has come below hearth in current months for failing to extra proactively police its Play Retailer for malicious apps, or for once-legitimate functions that later go rogue. This May 2023 story from Ars Technica a few previously benign display recording app that turned malicious after garnering 50,000 customers notes that Google doesn’t remark when malware is found on its platform, past thanking the skin researchers who discovered it and saying the corporate removes malware as quickly because it learns of it.

“The corporate has by no means defined what causes its personal researchers and automatic scanning course of to overlook malicious apps found by outsiders,” Ars’ Dan Goodin wrote. “Google has additionally been reluctant to actively notify Play customers as soon as it learns they have been contaminated by apps promoted and made obtainable by its personal service.”

The Ars story mentions one probably constructive change by Google of late: A safety measure obtainable in Android variations 11 and better that implements “app hibernation,” which places apps which were dormant right into a hibernation state that removes their beforehand granted runtime permissions.