Jupyter Notebooks

The Jupyter Notebook is an open-source web application that allows you to create and share documents that contain live code, equations, visualizations and narrative text. Uses include: data cleaning and transformation, numerical simulation, statistical modeling, data visualization, machine learning, and much more.


  • Docker CE : Docker Community Edition (CE) is ideal for developers and small teams looking to get started with Docker and experimenting with container-based apps.

Install Jupyter Notebook Server

You can simply download and run a docker image already created by the OTR Community. The Docker images are under the following account: https://hub.docker.com/u/cyb3rward0g. Look for the docker image names that start with jupyter-. If we want to download and run the jupyter-base image, you can do it with the following command:

$ docker run -p 8888:8888 cyb3rward0g/jupyter-base:latest

Download Mordor Dataset

Now that you have a jupyter notebook server up and running, we can download and decompress our dataset inside of the notebook. You can simply click on the first cell in your Jupyter Notebook interface and run the following command:

! curl -LJO https://raw.githubusercontent.com/OTRF/mordor/master/datasets/small/windows/lateral_movement/host/covenant_wmi_wbemcomn_dll_hijack.zip
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  605k  100  605k    0     0  3522k      0 --:--:-- --:--:-- --:--:-- 3522k

Next, decompress the specific mordor log file (Remember to run this in your Jupyter Notebook interface)

! unzip covenant_wmi_wbemcomn_dll_hijack.zip 
Archive:  covenant_wmi_wbemcomn_dll_hijack.zip
  inflating: covenant_wmi_wbemcomn_dll_hijack_2020-10-09173318.json

Explore as a Dataframe

Now, we can simply use python libraries such as Pandas to read our JSON file to a dataframe (Remember to run this in your Jupyter Notebook interface)

import pandas as pd
df = pd.read_json('covenant_wmi_wbemcomn_dll_hijack_2020-10-09173318.json')

Run a basic query to get Sysmon EventID 1 (ProcessCreate) events (Remember to run this in your Jupyter Notebook interface)


.assign(CommandLineLength = df['CommandLine'].str.len())
[(df['Channel'] == 'Microsoft-Windows-Sysmon/Operational') & (df['EventID'] == 1)]