وحدة 2
OPSEC prerequisites for detecting malware
آخر تحديث في: 23 أغسطس 2024
تعديل هذه الصفحة على GitHubوحدة 2
آخر تحديث في: 23 أغسطس 2024
تعديل هذه الصفحة على GitHubThis subtopic will allow the practitioner to ensure the security of the process and the individuals involved and implement a security policy within the computing environment the practitioner uses for malware detection.
After completing this subtopic, the practitioner should be able to ensure the confidentiality and integrity of data, which includes:
Operational security for detecting malware can be divided into concerns related to specific scenarios:
In many cases you will be handed a device and asked to inspect it for malware (or you may need to do this to your own device).
Be aware that in case the device is compromised, your activities may be monitored, which may impact risk and safety for your client. Any keystrokes including access to online accounts or communications may be captured. External storage devices such as hard drives or USB flash drives may become targets for transfer of malicious code, and any network connections may be used to further spread or exfiltrate malicious code.
Note as well that introducing analysis tools can trigger a ‘kill switch’ on some malware which has been designed to evade detection and analysis. In such cases, capturing a disk image and other forensic records may be necessary for further analysis. This is not covered in this learning path but is covered in Analyzing Malware.
If you suspect that a device is infected with malware, you should do as little as possible with it until you learn more about its status. For this reason, you should always use a device where you do not suspect any malware infections to handle any sensitive information.
If, for example, a person you are supporting suspects that their laptop or desktop might have been compromised, ask them to just use their mobile phone to communicate with you. It’s usually a good idea to shut down the potentially compromised laptop or desktop or at least disconnect it from the internet. If your beneficiary has tied their Signal, WhatsApp, and other accounts to the potentially compromised device, it might be a good idea to unlink those (doing so from a device that you do not suspect to be compromised) while the detection process is ongoing.
When going through the malware detection process, you might encounter links or files (either ordinary files or executable files) you are unsure about and which you suspect might be delivering malware payloads. If you are copying those links or files from a potentially compromised device to an analysis device, there’s always a risk that they could also infect your analysis device. In order to reduce the chances of this happening, we recommend:
For a deeper look at the topic, review the guide by Defensive Lab Agency on how to handle a potentially compromised device, in particular:
This last term on chain of custody refers to best practice in digital forensics and incident response to record the handling of a device in order to preserve evidence and allow evidence collected to be used in any potential legal proceedings. The linked article provides a good introduction to general-purpose best practices you can follow in case you are in a position to be handling evidence which might be used in a scenario with a higher burden of proof on evidence.
Set up a VM running REmnux, with the steps outlined in the Field Guide to incident response for civil society and media (chapter 6, starting on page 30).
After you have set up your REmnux VM, install and then connect to a reputable VPN. Make sure that your main system is either not connected to a VPN or to a different server than your REmnux instance. Ask your peer or mentor to send you a web bug canary token which will be opened just in REmnux, through a web browser of your choice. (If you are not yet familiar with canary tokens, check out this guide we created on how you could use them in security trainings.)
Which IP address did it trigger? What user agent?
Talk to your peer / mentor about what data stays on your VM and what doesn’t. If you ran a piece of malware in your VM which contacted a server, would this go through your VPN or home/ office network connection?
Intermediate guide - How to handle a potentially compromised device
FreeA step-by-step guide on how to handle devices with iOS or Android which you suspect might have malware on them prior to starting detection work
Virtual machine chapter of the Field Guide to incident response for civil society and media (chapter 6)
FreeAn introductory overview of how malware analysts can start working with virtual machines and an installation of the Linux distribution
Technical simulation with canary tokens
FreeA guide on how to use canary tokens, an offensive security tool, to simulate malware trackers. Can be very useful in teaching defenders what data can be easily exfiltrated
تهانينا على انتهاء وحدة 2!
وسم المربع لتأكيد اكتمالك والمتابعة إلى الوحدة التالية.
يوسم الوحدة الحالية كمكتملة ويحفظ التقدم للمستخدم.
لقد أكملت جميع الوحدات في مسار التعلم هذا.