Harvis is designed to automate your C2 Infrastructure.

Related tags

Text Data & NLPHarvis
Overview

Harvis

Harvis is designed to automate your C2 Infrastructure, currently using Mythic C2.

📌 What is it?

Harvis is a python tool to help you create multiple hauls for a specific operation during a red team engagement. It can automatically create your C2 machine, redirector machine, setup SSL, .htaccess configuration, firewall rules and more. Harvis also has the purpose of automating the job of the operator of identifying burned domains/redirectors that may be caught during the operation. After identifying the burned domains it provides the possibility of rotating the infrastructure, setting up different redirectors and atributing a different domain.

📌 How?

🔨 Harvis uses the Digital Ocean API to automate creation and deletion of droplets.

🔨 The Namecheap API is used to set DNS records to redirectors.

🔨 The API Void is used to constantly verify the state of the redirectors and check if anything is blacklisted.

Features

Harvis has several features to help you organize your available domains and redirector/C2 machines.

  • Namecheap Interaction - Harvis enables you to buy domains directly through the command line, as long as you have credits in your account.
  • Multiple Hauls - It is possible to create as many hauls as possible, each one having specific configurations.
  • Multiple C2 Profiles by Haul - Harvis allows you to create multiple C2 profiles by hauls: you could two HTTP listeners in one of the C2's and 3 in the other one, each of them listening on different ports.
  • Multiple Redirector Configuration - Each redirector can have a different .htaccess configuration, defined by the operator.
  • Customizable Firewall Rules - Harvis comes with default firewall rules for the redirectors and C2's, however, this feature is easily customizable.
  • Priority System - Each haul has a queue system, in a way that the new redirector will replace the blacklisted one with the domains in the queue.
  • Priority System - Automatically replace your droplets: Harvis identifies any blacklisted redirector and print the results to the operator. It allows the operator to create a temporary droplet to replace the blacklisted one. It does not configure the migration of any active agents, since the way the agent will be migrated/spawned to connect to the new domain might be a very personal decision in an engagement. After creating the temporary droplet, it allows you to migrate any active agents and kill the older redirector.
  • Priority System - If somehow the script crashes, all the information will be saved in the backup.py file. Restarting the script will recover all your infrastructure as it was.

Installation

git clone https://github.com/thiagomayllart/Harvis/
cd Harvis
pip3 install -r requirements

Running

python3 harvis.py

Configuring API Keys

Harvis can only be used with the proper API Keys from Digital Ocean, Namecheap, APIVoid.

These api keys should be added to the config.py file in the respective lines. Don't forget to setup your namecheap username in namecheap_username variable:

digital_ocean_token = ""
...
namecheap_key = ""
...
apivoid_key = ""
...
namecheap_username = ""
...

For more information regarding these API Keys, visit:

https://app.apivoid.com/ https://ap.www.namecheap.com/ https://cloud.digitalocean.com/

First Run

Before running the script, you should apply some modifications to the config file, which describes the configuration of your infrastructure:

  1. Modify the "names" variable.

This variable holds the names of each haul you want your infrastructure you have. Theses names should be applied in the next variables. You can have as many hauls you want. You can also specify which domain should already be configure to each haul. If you don't specify the domains, you will be asked to move domains to each haul in the first run. If you don't have any available domains in your namecheap API, you can buy it directly though Harvis. It is also possible to have more than one domain in each haul: the first one will be used in the redirector and the others will already be in the backup list for further infrastructure rotations. Example:

names = {"short":[],"long":[],"exploitation":[],"testing":[]}
names = {"short":["domain1.com"],"long":["domain2.com","domain3.com"],"exploitation":["domain4.com"],"testing":["domain5.com"]}
  1. Modify the "config_htaccess_dic" variable:

You can customize your htaccess rules for each haul in this variable. This variable is dictionary, so remember to add an htacces for each haul you added previously in the "names" variable like:

config_htaccess_dic = \
    {"short":"""
RewriteEngine On
RewriteCond %{REQUEST_URI} ^/({1})/?$ [NC]
RewriteRule ^.*$ https://{2}%{REQUEST_URI} [P]
RewriteRule ^.*$ http://{3}? [L,R=302]
""","long:"""
RewriteEngine On
RewriteCond %{REQUEST_URI} ^/({1})/?$ [NC]
RewriteRule ^.*$ https://{2}%{REQUEST_URI} [P]
RewriteRule ^.*$ http://{3}? [L,R=302]
""","exploitation":"""
RewriteEngine On
RewriteCond %{REQUEST_URI} ^/({1})/?$ [NC]
RewriteRule ^.*$ https://{2}%{REQUEST_URI} [P]
RewriteRule ^.*$ http://{3}? [L,R=302]
""","testing":"""
RewriteEngine On
RewriteCond %{REQUEST_URI} ^/({1})/?$ [NC]
RewriteRule ^.*$ https://{2}%{REQUEST_URI} [P]
RewriteRule ^.*$ http://{3}? [L,R=302]
"""
                       }

You can notice the presence of the fields: {1}, {2}, {3}. If you customize this variable, do not remove them. {1} are the parameters your Mythic agent will use to communicate to the C2 (it also allows customization). {2} is the IP address of your C2. {3} is the location the redirector will be redirecting (it also allows customization).

  1. Modify the "agent_profiles" variable:

This variable describes the HTTP parameters that your agent will use to communicate to your C2. Mythic allows setting these parameters during the creation of the agent, so, these values should match the ones you will be configurating the agent. The first one is the GET parameter and other one is the POST parameter. Also, remeber to once again add a configuration to each haul you created previously:

agent_profiles = {"short":{"URI":"data|index"},"long":{"URI":"q|id"}... ...
  1. Modify "domain_front_redirector" variable:

This variable holds the domain your redirectors will be redirecting anyone that tries to access. Add a configuration to each haul you created previously:

domain_front_redirector = {"short":"www.example.com","long":"www.example2.com"... ... ...

  1. Modify "c2_profiles" variable:

With the "c2_profiles" variable you can create different listener profiles for each Haul you created. The format is exactly the same as the JSON you may find when accessing Configuring a C2 Profile in Mythic. You can also have other profiles than HTTP, however, depending on the protocol used, it may be necessary to change firewall rules in the C2 or the redirector (further explained).

Example:


c2_profiles = {"short":[{"name":"HTTP","config":"""{
  "instances": [
  {
    "ServerHeaders": {
      "Server": "NetDNA-cache/2.2",
      "Cache-Control": "max-age=0, no-cache",
      "Pragma": "no-cache",
      "Connection": "keep-alive",
      "Content-Type": "application/javascript; charset=utf-8"
    },
    "port": 443,
    "key_path": "privkey.pem",
    "cert_path": "cert.pem",
    "debug": true
    }
  ]
}"""}],"long":[{"name":"HTTP","config":"""{
  "instances": [
  {
    "ServerHeaders": {
      "Server": "NetDNA-cache/2.2",
      "Cache-Control": "max-age=0, no-cache",
      "Pragma": "no-cache",
      "Connection": "keep-alive",
      "Content-Type": "application/javascript; charset=utf-8"
    },
    "port": 443,
    "key_path": "privkey.pem",
    "cert_path": "cert.pem",
    "debug": true
    }
  ]
}"""}] }
  1. Modify the "check_infra_state" variable:

This variable holds the value (in seconds) that will be used as the interval between each verification of blacklisted domains by APIVoid.

  1. Modify "ip_allowed_to_connect_c2":

Replace it with the IP you will be using as the proxy to connect to your Mythic C2 panel. You can use your public IP, but it is not recommended.

  1. Replace "username":

Replace it with the username you will be using during the engagement. This variable is used to tag each droplet created in digital ocean, making the distinction between the droplets of each operator easier. It also avoids that the tool erases the droplet of another user (in case you guys are using the same API Keys).

  1. Modify Firewall Rules (OPTIONAL):

If you want to modify the firewall rules for the C2/redirector, you will find them respectively at:

C2: C2_setup.py: function firewall_rules Redirector: redirect_setup.py: firewall_rules

Important

Harvis whitelists the IP of the machine you are deploying it as being able to access your Mythic C2 panel. It is highly recommended to use a VPC to deploy Harvis.

Owner
Thiago Mayllart
- Information Security Researcher / RedTeam
Thiago Mayllart
NLP codes implemented with Pytorch (w/o library such as huggingface)

NLP_scratch NLP codes implemented with Pytorch (w/o library such as huggingface) scripts ├── models: Neural Network models ├── data: codes for dataloa

3 Dec 28, 2021
Code to reprudece NeurIPS paper: Accelerated Sparse Neural Training: A Provable and Efficient Method to Find N:M Transposable Masks

Accelerated Sparse Neural Training: A Provable and Efficient Method to FindN:M Transposable Masks Recently, researchers proposed pruning deep neural n

itay hubara 4 Feb 23, 2022
AI Assistant for Building Reliable, High-performing and Fair Multilingual NLP Systems

AI Assistant for Building Reliable, High-performing and Fair Multilingual NLP Systems

Microsoft 37 Nov 29, 2022
Transcribing audio files using Hugging Face's implementation of Wav2Vec2 + "chain-linking" NLP tasks to combine speech-to-text with downstream tasks like translation and summarisation.

PART 2: CHAIN LINKING AUDIO-TO-TEXT NLP TASKS 2A: TRANSCRIBE-TRANSLATE-SENTIMENT-ANALYSIS In notebook3.0, I demo a simple workflow to: transcribe a lo

Chua Chin Hon 30 Jul 13, 2022
Active learning for text classification in Python

Active Learning allows you to efficiently label training data in a small-data scenario.

Webis 375 Dec 28, 2022
pysentimiento: A Python toolkit for Sentiment Analysis and Social NLP tasks

A Python multilingual toolkit for Sentiment Analysis and Social NLP tasks

297 Dec 29, 2022
Tool to add main subject to items on Wikidata using a WMFs CirrusSearch for named entity recognition or a manually supplied list of QIDs

ItemSubjector Tool made to add main subject statements to items based on the title using a home-brewed CirrusSearch-based Named Entity Recognition alg

Dennis Priskorn 9 Nov 17, 2022
Code to reproduce the results of the paper 'Towards Realistic Few-Shot Relation Extraction' (EMNLP 2021)

Realistic Few-Shot Relation Extraction This repository contains code to reproduce the results in the paper "Towards Realistic Few-Shot Relation Extrac

Bloomberg 8 Nov 09, 2022
The Internet Archive Research Assistant - Daily search Internet Archive for new items matching your keywords

The Internet Archive Research Assistant - Daily search Internet Archive for new items matching your keywords

Kay Savetz 60 Dec 25, 2022
A large-scale (194k), Multiple-Choice Question Answering (MCQA) dataset designed to address realworld medical entrance exam questions.

MedMCQA MedMCQA : A Large-scale Multi-Subject Multi-Choice Dataset for Medical domain Question Answering A large-scale, Multiple-Choice Question Answe

MedMCQA 24 Nov 30, 2022
Create a machine learning model which will predict if the mortgage will be approved or not based on 5 variables

Mortgage-Application-Analysis Create a machine learning model which will predict if the mortgage will be approved or not based on 5 variables: age, in

1 Jan 29, 2022
Hierarchical unsupervised and semi-supervised topic models for sparse count data with CorEx

Anchored CorEx: Hierarchical Topic Modeling with Minimal Domain Knowledge Correlation Explanation (CorEx) is a topic model that yields rich topics tha

Greg Ver Steeg 592 Dec 18, 2022
Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities

Hiring We are hiring at all levels (including FTE researchers and interns)! If you are interested in working with us on NLP and large-scale pre-traine

Microsoft 7.8k Jan 09, 2023
A pytorch implementation of the ACL2019 paper "Simple and Effective Text Matching with Richer Alignment Features".

RE2 This is a pytorch implementation of the ACL 2019 paper "Simple and Effective Text Matching with Richer Alignment Features". The original Tensorflo

286 Jan 02, 2023
NVDA, the free and open source Screen Reader for Microsoft Windows

NVDA NVDA (NonVisual Desktop Access) is a free, open source screen reader for Microsoft Windows. It is developed by NV Access in collaboration with a

NV Access 1.6k Jan 07, 2023
Transformer Based Korean Sentence Spacing Corrector

TKOrrector Transformer Based Korean Sentence Spacing Corrector License Summary This solution is made available under Apache 2 license. See the LICENSE

Paul Hyung Yuel Kim 3 Apr 18, 2022
The source code of "Language Models are Few-shot Multilingual Learners" (MRL @ EMNLP 2021)

Language Models are Few-shot Multilingual Learners Paper This is the source code of the paper [Arxiv] [ACL Anthology]: This code has been written usin

Genta Indra Winata 45 Nov 21, 2022
A Fast Command Analyser based on Dict and Pydantic

Alconna Alconna 隶属于ArcletProject, 在Cesloi内有内置 Alconna 是 Cesloi-CommandAnalysis 的高级版,支持解析消息链 一般情况下请当作简易的消息链解析器/命令解析器 文档 暂时的文档 Example from arclet.alcon

19 Jan 03, 2023
FastFormers - highly efficient transformer models for NLU

FastFormers FastFormers provides a set of recipes and methods to achieve highly efficient inference of Transformer models for Natural Language Underst

Microsoft 678 Jan 05, 2023