Vulnerability Patch Management with LLM

An AI-powered system that automates vulnerability analysis and patch recommendations using Large Language Models (LLMs). Parses Nmap scan results, prioritizes risks, and generates actionable remediation steps.

📂 Repository Structure

├── app.py # Main FastAPI/Flask application (backend server) ├── model.py # LLM integration & vulnerability analysis logic ├── utils.py # Helper functions (Nmap XML parsing, etc.) ├── requirements.txt # Python dependencies ├── templates/ # HTML templates for web interface (if applicable) ├── pycache/ # Python cache directory (ignored in Git) └── .dist/ # Build/distribution files (optional)

text

🚀 Features

• Automated Scan Analysis: Parse Nmap XML outputs to identify outdated services.
• LLM-Powered Recommendations: GPT-4/Llama 3 suggests patches based on CVE databases.
• Prioritization Dashboard: Ranks vulnerabilities by severity (Critical/High/Medium).
• Self-Healing: Optional auto-remediation for low-risk patches (e.g., Ansible scripts).

⚙️ Setup

1. Clone the repo: Install dependencies:

bash pip install -r requirements.txt Add your LLM API key in config.py (or environment variables).

🖥️ Usage python

Analyze Nmap scan

from utils import parse_nmap from model import get_patch_recommendations

scan_data = parse_nmap("scan_results.xml") recommendations = get_patch_recommendations(scan_data)