In this article, we learn how to use Commix from scratch by using all the basic commands and going all the way to the advanced ones.
Table of Content
- Introduction to command injection
- Introduction to Commix
- Working of Commix
- Types of Commix
- Requirements
Introduction to Command Injection
Command injection is also known as shell injection or OS injection. Command injection is one of the top 10 OWASP vulnerability. it’s an attack in which arbitrary commands of a host OS are executed through a vulnerable application. Such an attack is possible when a web application sends unsafe user data to the system shell. This user data can be in any form such as forms, cookies, HTTP headers, etc. Mostly the vulnerability command injection rises due to insufficient input validation. In this attack, the default functionality of the application is extended by an attacker who then executed the system commands with injecting code which makes it different from code injection.
The process of command injection was accidentally discovered in 1997 by a programmer in Norway. This accident led to the deletion of web pages of a site. SQL command injection is the most popular form of command injection. Through this attack, an attacker adds SQL code to the input box in order to gain access. Web applications are compulsory for such attacks as we communicate with the underlying OS via such web applications.
Introduction to Commix
Commix tool is automated for exploiting the vulnerability of command injection in web applications. This tool is written in python which means it is compatible with Linux, windows and mac. It comes pre-installed in Kali Linux, BlackArch and Parrot OS. This tool makes it very easy to find vulnerabilities related to command injection and then further exploit them. The user-friendliness of commix makes it very convenient for everyone, such as web developers, pen testers or security researchers, to use it.
It provides a user with a lot of options such as including the ability to specify parameters that you need to connect to host, enumeration of a victim, accessing files and their modification along with an offline mode. Hence, it’s a pretty useful asset in order to exploit command injection vulnerability.
Working of Commix
Commix has various command options which you can use to find and connect with the target application. Few of the options target URL is via data strings, HTTP headers, cookies and authentication parameters. There are various enumerations options present too. Commix supports two command injection techniques i.e. result-based command injection technique and blind command injection technique. Result based command injection is the once where commands are reflected back to the attacker in the web application. Whereas blind command injection technique is persuading when the response is not reflected on a web application.
Types of Command Injection in Commix
Result based command injection
This type of injection attack will let you deduce the result of the injected command through the result of a web application. It is further divided into two categories :
- Classic result based injection : This is the most commonly used type of command injection and is the simplest of all. In this, several common operators are used which either links genuine commands with the injected ones or exclude the initial commands altogether and goes ahead to execute the injected ones only. this further divides into 3 categories i.e. Shellshock, ICMP exfiltration, DNS exfiltration.
- Eval-Based technique : This technique is used where the targeted web application is vulnerable to eval() function. This eval function is used to execute the peculiar code that is defined in the said function during run time.
Blind Command Injection
The way that the data is retrieved after the execution of injected shell command is the main difference between the working of both the types. In the case where web application does not give any result back to the attacker; blind command injection is used. There are further two types of blind command injection :
- Time-based Technique : Using this technique will delay the time of the execution of a command that is injected. By checking how much time the application took to revert will able the attacker to determine whether the command is executed successfully or not.
- File-based Technique : If you are not able to determine the result of the web application through its reaction, then this technique comes in handy as it will allow you to write the set of commands that are to be injected in the file accessible to the attacker. Working of this technique is similar to that of result based technique.
Requirements
- DVWA
- PentestLab
- Kali Linux
- Commix Tool
Some of the practical we will do in pentestlab for Linux and others we will perform on DVWA for windows. Let’s start with the practical of commix. First, we will use help command to check all the options that we use to exploit target via commix.
Now let’s try and get a commix session using the URL. For this, use the URL that is vulnerable to the command injection, here, we will pentesterlab’s URL as shown in the image below :
Use the following command to have a commix session through URL :
And so, with this command you get a commix(os_shell) as shown in the image below :
Now, let’s use the batch command to have the commix session by default and for this use the following command :
Now, as you can see in the above image, we have directly entered the session by default. Now, to get all the basic information about the target use the following command :
As shown in the image above, using the above command will give you all the basic information about the target. Next, with the help of the following command we can know the current user of target :
As a result of the above command, you can see in the above that the current user is www-data. Now, we can also find out the hostname by using the following command :
Hence, the host is Debian. Now, the question is how we can determine whether the out target is rooted or not. So, for this, we have the following command :
This way we found that our target was not rooted. Use the following command to have the information about the system :
Upon executing the above command, we now have the system information. To have information about users, use the following command :
And this way, we have a list of all the users. Next command is used to know about the admin of the system :
And again you can see that the target is not the admin. Now to read the contents of a desired file, we can use the following command :
And the results of the above command are shown in the image above. Our next practicals are performed on DWVA (windows environment)
Captured the cookies of the submitted request using BurpSuite.
Now, we use this content of cookie to validate our session using the parameters ‘—cookie’ and ‘—data’. These two parameters are used to send data string in order to exploit the POST method and to validate our session simultaneously. For this, use the following command :
With the help of above the command, we will directly have a session as shown in the image above. Now, we will create a malware file using msfvenom. Type the following command to generate your malware :
Now, we will use the above file and upload it in our target by using the following command :
Now, the above command will upload and run our malware in the target machine. You can use multi/handler to get a session and for this use the following set of commands :
And this way, as shown in the image, you will have a meterpreter session. This is how you can use commix, a third party automated tool, to your advantage.