Introduction
Welcome to the official documentation for the VULNRIX Autonomous Security Platform.
Community Edition
The full platform — completely free. Autonomous AI agents, Playwright deep crawler, Nuclei zero-day fuzzing, and continuous asset monitoring. No license key required.
Powered by Ollama
VULNRIX uses Ollama to run AI models 100% locally on your machine. No internet. No API keys. No cloud. Download Ollama first before running VULNRIX.
System Overview
The VULNRIX platform is designed to operate as a completely self-contained offensive security unit. It requires no internet connection to function once set up — all AI inference runs locally via Ollama.
Core Components
- Orchestrator: The central AI brain — plans the attack strategy and coordinates agents.
- Recon Agent: Playwright-powered deep crawling + passive subdomain enumeration.
- Exploit Agent: Nuclei-based zero-day fuzzing and logic flaw analysis.
- Continuous Monitor: Celery Beat scheduled asset diffing — finds new targets automatically.
- Report Engine: Structured vulnerability report generation ready for bug bounty submission.
Quick Start
VULNRIX is designed to be as easy to run as Ollama itself. Three steps and you are hunting bugs:
# Step 1 — Install Ollama (runs your AI models locally) # Download from: https://ollama.com/download # Step 2 — Pull a model (e.g. llama3 or mistral) ollama pull llama3 # Step 3 — Clone VULNRIX and run setup git clone https://github.com/vulnrix/AI-HITL cd AI-HITL bash setup.sh # Step 4 — Start the platform python manage.py runserver # Open http://localhost:8000 in your browser
Hardware Requirements
Minimum 16GB RAM recommended for running local AI models via Ollama. 32GB+ RAM is ideal for running multiple concurrent scanning agents. No GPU required — CPU inference works great for bug bounty workflows.