Skip to content

gmh5225/ORHUS

 
 

Repository files navigation

🔐 ORHUS (Offensive Reasoning & Heuristic Unified System)

Offensive Reasoning & Heuristic Unified System

Autonomous AI-Powered Penetration Testing Agent

ORHUS is an industry-grade, red-teaming agent designed to automate the strategic phases of a penetration test within authorized lab environments. It combines high-fidelity reconnaissance with an advanced AI decision engine to simulate the logic and tactics of an expert security consultant.

⚡ Technical Core

  • Phase 1: Reconnaissance — High-fidelity service and version mapping using Nmap.
  • Phase 2: Vulnerabilities — Automated CVE matching and CVSS scoring.
  • Phase 3: Exploitation — Metasploit integration and AI-guided custom exploit logic.
  • Phase 4: Reporting — Automated, high-fidelity HTML/PDF reports (Powered by VOIDSEC Labs).
  • Phase 5: Memory — Persistent session tracking and cross-target intelligent learning.

🛠️ Security Framework

Feature Description
Orchestrator Central autonomous loop managing the test flow.
Brain Multi-AI Decision Engine (Gemini & Grok).
Memory Global intelligence database for successful exploit patterns.
Safe Mode Hard-coded RFC 1918 and platform-specific IP restrictions.

🚀 Quick Install

Kali Linux / Parrot OS

git clone https://github.com/krithick-rk/ORHUS.git
cd ORHUS
sudo ./setup/install.sh
cp .env.example .env
# Add your API keys to .env
python3 main.py

Windows

git clone https://github.com/VOIDSEC-LABS/ORHUS.git
cd ORHUS
setup\install_windows.bat
copy .env.example .env
# Add your API keys to .env
python main.py

⚠️ Disclaimer

ORHUS is for authorized security testing only. Unauthorized use against systems without explicit written permission is illegal and strictly prohibited.


© 2026 VOIDSEC Cybersecurity Solutions | voidsec.info

About

ORHUS (Offensive Reasoning & Heuristic Unified System) is an AI-powered autonomous penetration testing agent that simulates real pentester decision-making. Phase 1 includes reconnaissance automation, AI-driven analysis (Gemini/xAI), and intelligent next-step recommendations in a safe lab-only environment.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • Python 86.1%
  • HTML 7.8%
  • CSS 2.4%
  • Shell 2.3%
  • Batchfile 1.4%