Nowadays Iran’s Cybersecurity capabilities are under microscope, many news sites, gov. agencies and security experts warn about a possible cybersecurity infiltration from Iranian government and alert to increase cybersecurity defensive levels. Today I want to share a quick and short study based on cross correlation between MITRE ATT&CK and Malpedia about some of the main threat actors attributed to Iran. The Following sections describe the TTPs (Tactics, Techniques and Procedures) used by some of the most influential Iranian APT groups. Each section comes with a main graph which is built by scripting and which comes without legend, so please keep in mind while reading that: the red circles represent the analyzed threat actors, the green circles represent threat actor’s used techniques, the blue circles represent the threat actor’s used Malware and the black circles represent the threat actor’s used tool sets.
According to Malpedia: “OilRig is an Iranian threat group operating primarily in the Middle East by targeting organizations in this region that are in a variety of different industries; however, this group has occasionally targeted organizations outside of the Middle East as well. It also appears OilRig carries out supply chain attacks, where the threat group leverages the trust relationship between organizations to attack their primary targets.” The threat actor uses opensource tools such as Mimikatz and laZagne, common sysadmin toolset available on Microsoft distribution or sysinternals such as: PsExec, CertUtil, Netstat, SystemInfo, ipconfig and tasklist. Bonupdater, Helminth, Quadangent and PowRuner are some of the most sophisticated Malware attributed to OilRig and analyzed over the past few years. Techniques (green) are mainly focused in the lateral movements and in getting persistence on the victim infrastructure; few of them involved exploiting or 0days initiatives.
Those observations would suggest a powerful group mostly focused on staying hidden rather than getting access through advanced techniques. Indeed no 0days or usage of advanced exploits is found over the target infrastructure. If so we are facing a state-sponsored group with high capabilities in developing persistence and hidden communication channels (for example over DNS) but without a deep interest in exploiting services. This topic would rise a question: OilRig does not need advanced exploiting capabilities because it is such a simple way to get into a victim infrastructure ? For example by using: user credential leaks, social engineering toolkits, targeted phishing, and so on and so forth or is more on there to be discovered ?
According to MITRE: “MuddyWater is an Iranian threat group that has primarily targeted Middle Eastern nations, and has also targeted European and North American nations. The group’s victims are mainly in the telecommunications, government (IT services), and oil sectors.” Currently we have few artifacts related to MuddyWater (‘Muddy’), indeed only Powerstats backdoor is actually attributed to it. Their attack are typically “hands driven”, which means they do not use automation lateral movement but they prefer to use opensource tools or sysinternal ones to deliberately move between target network rather than running massively exploits or scanners.
Once landed inside a victim machine Muddy looks for local credentials and then moves back and forward by using such a credentials directly on the network/domain controllers. According to MITRE techniques (green) MuddyWater to take an entire target-network might take few months but the accesses are quite silent and well obfuscated. Again it looks like we are facing a group which doesn’t need advanced exploitation activities but rather than advanced IT knowledge in order to move between network segments and eventual proxies/nat.
According to MITRE: “APT33 is a suspected Iranian threat group that has carried out operations since at least 2013. The group has targeted organizations across multiple industries in the United States, Saudi Arabia, and South Korea, with a particular interest in the aviation and energy sectors.” Analyzing the observed TTPs we might agree that this threat actor looks very close to MuddyWater. If you take a closer look to the Muddy Graph (in the previous dedicated section) and APT33 graph (following) you will see many similarities: many tools are shared, many techniques are shared and even artifacts Powerstats (Muddy) and Powertron (APT33) share functions and a small subset of code (even if they have different code bases and differ in functionalities). We have more information about APT33 if compared to MuddyWatter, but similarities on TTPs could induce an avid reader to think that we might consider APT33 as the main threat actor while MuddyWater a specific ‘operation’ of the APT33 actor.
But if you wonder why I decided to keep them separated on such personal and preliminary analysis you could find the answer in the reason in why they do attack. APT33 showed destruction intents by using Malware such as shamoon and stoneDrill, while Muddy mostly wants to “backdooring” the victims.
According to MITRE: “CopyKittens is an Iranian cyber espionage group that has been operating since at least 2013. It has targeted countries including Israel, Saudi Arabia, Turkey, the U.S., Jordan, and Germany. The group is responsible for the campaign known as Operation Wilted Tulip.” CopyKittens threat actor actually differ from the previous ones. First of all we see the usage of CobaltStrike, which is an autonomous exploiting system (well actually is much more, but let me simplify it). Cobalt and Empire (a post exploitation framework) taken together would allow the attacker to automate lateral movement. Which is a damn different behavior respect to previous actors. CopyKittens would make much more noise inside an attacked network and would be easier to detect if using such automation tools, but on the other hand they would be much more quick in reaching their targets and run away.
One more characteristic is the “code signing”. While in OilRig, MuddyWater and APT33 we mostly observed “scripting” capabilities, in CopyKittens we are observing most advanced code capabilities. Indeed code signing is used on Microsoft Windows and IOS to guarantee that the software comes from known developer and that it has not been tampered with. While a script (node, python, AutoIt) could be attribute to IT guys as well as developers, developing more robust and complex software ( such as: java, .net, c++, etc) is a skill typically attributed to developers. This difference could be significant in suspecting a small set of different people working on CopyKittens.
According to MITRE: “Cleaver is a threat group that has been attributed to Iranian actors and is responsible for activity tracked as Operation Cleaver.  Strong circumstantial evidence suggests Cleaver is linked to Threat Group 2889 (TG-2889). ” We have a few information about this group, and as you might see there are few similarities. The usage of Mimikatz could be easily adopted for credential dumping, while TinyZBot is a quite interesting tool since it mostly implements spying capabilities without strong architectural design or code execution or data exfiltration.
Just like Charming Kitten (which is not included into this report since it is a quite ongoing mistery even if a great report from Clear Sky is available), Cleaver is a threat group that is responsible of one of the first most advanced and silent cyber attack attributed to Iran known until now (OpCleaver, by Cylance). Cleaver attack capabilities are evolved over time very quickly and, according to Cylance, active since 2012. They look like to have infiltrated some of the world economic powers (ref: here) such as: Canada, China, England, France, Germany, India, Israel, Kuwait, Mexico, Pakistan, Qatar, Saudi Arabia, South Korea, Turkey, United Arab Emirates, and the United States. In the very first page of the OpCleaver report, the author writes that Cleaver is one of the most advanced threat actors ever. Even if I might agree with Cylance, I personally do not have such evidences so far, so I personally cannot compare Cleaver threat actor to the previus ones.
Here comes the fun ! How about taking all these graphs and compare them ? Common references would highlight similarities, scopes and common TTPs and fortunately we might appreciate them in the following unique network diagram. You might spend over 20 minutes to check details on the following graph and I might decide to write an essay over it, but I will not do it :D, I’d like focus on few but important thoughts.
The iper-connection between the analyzed groups (take a look to the following graph) could prove that those teams are really linked together. They share Techniques, Procedures, Tools and Infection Artifacts and everything we might observe looks like belonging with a unique meta-actor. We might agree that the meta-actor would be linked to the sponsorship nation and we might decide to consider some of those groups as operations. In other words we might consider an unique group of people that teams up depending of the ongoing operation adopting similar capabilities and tool sets.
OilRig and APT33 are the most known groups attributed to Iran, they share many tools but they clearly have two different intent and two different code bases (writing about Malware). CopyKittens, for example, have been clustered more closed to APT33 while Muddywater looks like clustered straight at the middle of them. But if we closely analyze the purposes and the used Malware we might agree in aggregating Muddy close to APT33, actually the weight of shared code should be heavier compared to common tools or common techniques, but I did not represent such a detail into graphs.
However two different ‘code experience’ are observed. The first one mostly focused on scriptting (node, python, autoIT) which could underline a group of people evolving from IT department and later-on acquiring cyersecurity skills, while the second observed behavior is mostly oriented on deep development skills such as for example: Java, .NET and C++. On MuddyWater and APT33 side, the usage of scripting engines, the usage of powershell, and the usage of Empire framework tighten together, plus the lack of exploiting capabilities or the lack in developing sophisticated Malware could bring the analyst to think that those threat actors hit their target without the need of strong development capabilities. On the other hand OilRig, Cleaver and CopyKitten looks like to have more software developing skills and looks to be mostly focused on stealth operations.
In this post I wrote a preliminary and personal analysis of threat actors attributed by the community to Iran, comparing TTPs coming from MITRE and relations extracted from Malpedia. The outcome is a proposal to consider the numerous groups (OilRig, APT33, MuddyWater, Cleaver, etc..) as a primary meta-threat-actor and dividing them by operations rather real group.
Original Post published on Ramilli’s blog:
About the author: Marco Ramilli, Founder of Yoroi
I am a computer security scientist with an intensive hacking background. I do have a MD in computer engineering and a PhD on computer security from University of Bologna. During my PhD program I worked for US Government (@ National Institute of Standards and Technology, Security Division) where I did intensive researches in Malware evasion techniques and penetration testing of electronic voting systems.
I do have experience on security testing since I have been performing penetration testing on several US electronic voting systems. I’ve also been encharged of testing uVote voting system from the Italian Minister of homeland security. I met Palantir Technologies where I was introduced to the Intelligence Ecosystem. I decided to amplify my cybersecurity experiences by diving into SCADA security issues with some of the biggest industrial aglomerates in Italy. I finally decided to found Yoroi: an innovative Managed Cyber Security Service Provider developing some of the most amazing cybersecurity defence center I’ve ever experienced! Now I technically lead Yoroi defending our customers strongly believing in: Defence Belongs To Humans