DJSwiss / Neural-Evaluator

A neural network evaluating the success of automated penetration testing

NEURAL EVALUATOR MICROSERVICE

This is a wrapping microservice API for using the neural network for evaluating attack vectors on a website. It uses Python Flask for the API and PyTorch for the neural network.

This project is a cooperation of C&M of the Institute for Technology in Karlsruhe, Germany and IC-Consult. It was created as part of the master thesis of Cedric Mössner. You can see more information in the thesis paper found here:https://drive.google.com/open?id=1Gqq4D6mqnqFaYFtwZKSPALJu1LcqcU8z If you still need more information, please feel free to contact me at [e-mail].

USAGE with DOCKER

If not already done, install docker from here

Use the Dockerfile to create a docker image:

docker build -t evaluatormicroservice .
docker run -p 1337:1337 evaluatormicroservice

USAGE without DOCKER

Get Python 3 and Pip. Then do:

$ pip -r requirements.txt

On Linux then go ahead and install PyTorch for CPU:

$ pip3 install http://download.pytorch.org/whl/cpu/torch-0.4.1-cp36-cp36m-linux_x86_64.whl 
$ pip3 install torchvision

or follow the instructions at PyTorch.org

then you can run it with

$ python app.py