wavecrack: Wavestone’s web interface for password cracking with hashcat

wavecrack

Wavecrack

A user-friendly Web interface to share an hashcat cracking box among multiple users with some pre-defined options.

Outline

  • This Web application can be used to launch asynchronous password cracks with hashcat.
  • The interface tries to be as user-friendly as possible and facilitates the password cracking method choice and to automate the succession of various attack modes.
  • It also displays statistics regarding the cracked passwords and allows to export the cracked password list in CSV.
  • The application is designed to be used in a multi-user environment with a strict segregation between the cracking results of different users: the user authentication can be done through an LDAP directory or basic auth.

Installation

  • Install the RabbitMQ server and python-ldap requirements
    $ apt-get install libsasl2-dev libldap2-dev libssl-dev rabbitmq-server
  • Install the python requirements
    $ pip install -r requirements.txt
  • Create a cracker/app_settings.py configuration file from the cracker/app_settings.py.example file and notably edit the Mandatory settings section:
    • The path of hashcat
    • The RabbitMQ connection string: by default, the guest/guest account is used. Be sure to harden your installation
    • The path of the SQLite database
    • The path of the hashcat rules
    • The path of the wordlists
    • The LDAP parameters:
      • IP address
      • port
      • LDAP database for the users
      • Base DN
  • Initialize the local database linked in the cracker/app_settings.py configuration file
    $ sqlite3 base.db < base_schema.sql
  • Start the RabbitMQ server
    $ sudo service rabbitmq-server start
  • Start Celery from the application folder
    $ celery worker -A cracker.celery
  • Launch the Flask Web server
    • Directly from the server.py file: this mode is not suitable for production purpose
      $ python server.py
    • With a wsgi script: an example of app.wsgi.example is provided
    • Similarly, supervisorctl can be used to manage celery with a configuration file example in supervisorcelery.conf.example
  • In order to stop the cracks after a certain amount of time, you can use the provided cron script.
  • If you want to update the list of hashes supported, you can use the dedicated script which will parse hashcat’s wiki and generate an updated hashcat_hashes.py. To do so, you need to have BeautifulSoup installed on your system.

Finally, if you don’t want to setup your own VM, you can use the Docker-based process described in the docker folder.

Usage

Wavecrack can be used to do the following:

  • Add new password hashes, choose the attack mode and the crack duration
  • View the past and current cracks for your user with statistics and graphs
  • View the overall load of the platform
  • Upload a password-protected file and extract its hash

The attack modes are followed in the order they are displayed on the hash submit form.
It is also possible to stop a crack. However, every cancellation is final.
A limit to the number of concurrent cracks can be defined in the settings in order not to reduce the current cracks performance.

  • The homepage The homepage
  • Adding a hash to crack Adding an hash to crack
  • Seeing the results and some stats Seeing the results and some stats

Source: https://github.com/wavestone-cdt/