Skip to main content

AI-Powered Cybersecurity Copilot for Kali Linux — Think Claude Code, but built for hackers.

Project description

MundiX CLI

AI-Powered Cybersecurity Copilot for Kali Linux

MundiX CLI is an intelligent terminal assistant designed for penetration testers, red teamers, and security researchers. It connects to specialized cybersecurity AI models (DeepHat, WhiteRabbitNeo) via HuggingFace, and features a collective memory system that learns from all users.

Think of it as Claude Code or GitHub Copilot, but built specifically for hackers.


Features

Feature Description
AI Chat Interactive conversation with uncensored cybersecurity AI models
Command Execution Extract and execute commands directly from AI responses
Collective Memory Local + shared memory that improves with every user interaction
Multi-Model Switch between DeepHat, WhiteRabbitNeo, and other models
System Awareness AI knows your OS, tools, and current directory
Safety Checks Dangerous commands are flagged before execution
Streaming Real-time token streaming for fast responses
One-Shot Mode Quick queries without entering interactive mode

Quick Install (Kali Linux)

curl -sSL https://raw.githubusercontent.com/k0k4/mundix-cli/main/install.sh | bash

Manual Install

git clone https://github.com/k0k4/mundix-cli.git
cd mundix-cli
pip3 install -r requirements.txt
export HF_TOKEN="your_huggingface_token"
python3 mundix_cli.py

Install via pip

pip3 install -e .
mundix

Get your free HuggingFace token at: https://huggingface.co/settings/tokens


Usage

Interactive Mode

mundix

One-Shot Query

mundix "how to scan for open ports on 10.0.0.1"

Execute and Analyze

mundix --exec "nmap -sV 10.0.0.1"

Choose a Model

mundix --model wrn-qwen-7b

Available Models

Key Model Description
deephat-7b DeepHat V1 7B Cybersecurity & DevOps specialist (default)
wrn-qwen-7b WhiteRabbitNeo 2.5 Qwen 7B Offensive/Defensive cybersecurity
wrn-llama-8b WhiteRabbitNeo Llama 8B v2 General cybersecurity assistant

CLI Commands

Inside the interactive shell, use these slash commands:

Command Description
/help Show help message
/models List available AI models
/model <name> Switch to a different model
/memory Show memory statistics
/sync Sync collective memory
/clear Clear conversation history
/system Show system information
/history Show command execution history
/exec <cmd> Execute a command directly
/analyze Send last command output to AI
/exit Exit MundiX CLI

Collective Memory

MundiX CLI features a unique collective memory system:

How it works: Every useful interaction (questions, commands, solutions) is saved locally. High-quality entries (upvoted by the user) are anonymized and shared with the MundiX collective memory. When you ask a question, the CLI first searches the collective memory for similar past solutions before querying the AI model, saving tokens and leveraging community knowledge.

Privacy: Only anonymized command patterns and solutions are shared. No personal data, IPs, or target information is ever transmitted.


Project Structure

mundix-cli/
  mundix_cli.py          # Main CLI application (REPL, UI, slash commands)
  ai_provider.py         # HuggingFace API integration (OpenAI-compatible)
  memory.py              # Collective memory system (local + sync)
  system_integrator.py   # Command execution and system interaction
  install.sh             # Quick installer for Kali Linux
  requirements.txt       # Python dependencies
  setup.py               # pip install support
  test_modules.py        # Module test suite
  LICENSE                # MIT License

Requirements

  • Python 3.8+
  • Linux (optimized for Kali Linux)
  • HuggingFace API token (free tier works)

Roadmap

  • TUI dashboard with Rich panels (scan results, AI chat side by side)
  • Plugin system for custom tools integration
  • Collective memory API server for real-time community sync
  • Voice input/output for hands-free operation
  • Auto-reconnaissance mode (full automated recon pipeline)
  • Integration with Metasploit RPC and Burp Suite API

License

MIT License - MundiX


Built with purpose by MundiX. Happy Hacking.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

mundix_cli-0.9.6-py3-none-any.whl (234.4 kB view details)

Uploaded Python 3

File details

Details for the file mundix_cli-0.9.6-py3-none-any.whl.

File metadata

  • Download URL: mundix_cli-0.9.6-py3-none-any.whl
  • Upload date:
  • Size: 234.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.2

File hashes

Hashes for mundix_cli-0.9.6-py3-none-any.whl
Algorithm Hash digest
SHA256 8e6803252b1a10e09d4cca075cdf18220816f02dc52272d2ecb90fe08ae00418
MD5 a3ee83fbeda35a6b4dde74ebd671d5e6
BLAKE2b-256 8a7c08c3524f1f4333c5cd758f66ec53cff64d435bc6993f979b05719a650944

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page