LVing - Vulnerability Detection Framework

A comprehensive static analysis tool built for identifying memory safety and concurrency vulnerabilities in programming languages leveraging the LLVM compiler framework, with a particular interest in the Rust language, through graph-based visualization and analysis.

Overview

LVing is a static analysis framework for detecting memory safety and concurrency vulnerabilities in Rust programs through graph-based program representation. This tool constructs Code Property Graphs (CPGs) from LLVM Intermediate Representation, enabling systematic analysis of low-level memory operations and vulnerability patterns. Built on the Fraunhofer Institute's CPG library, LVing provides a scalable platform for automated vulnerability detection and interactive program exploration via Neo4j graph visualization.

---

Prerequisites

• Docker & Docker Compose: Install Docker (includes Docker Compose)

---

Key Files:

• docker-compose.yml: Orchestrates backend, frontend, and Neo4j services
• .env: Environment variables (Neo4j credentials)

---

Environment Variables

Create a .env file in the project root with your configuration:

# Required: Neo4j password
NEO4J_PASSWORD=your_password_here

# IP address where Neo4j and backend services will be accessible
# For local development: localhost
# For remote deployment: your server's external IP
NEO4J_IP=your_neo4j_ip_here

# Backend host configuration (defaults to NEO4J_IP if not specified)
BACKEND_HOST=your_backend_ip_here
BACKEND_PORT=8000

# Neo4j browser host (what users see in browser, usually same as NEO4J_IP)
NEO4J_BROWSER_HOST=your_neo4j_ip_here

---

Getting Started

1. Setup Environment

git clone <repository-url>
cd LVing
cp .env.example .env
# Edit .env with your configuration:
# - Set NEO4J_PASSWORD to a secure password
# - Set NEO4J_IP to your server's IP (or localhost for local development)
# - Optionally customize BACKEND_HOST if different from NEO4J_IP

2. Deploy Services

docker compose build
docker compose up

3. Access Services

• Frontend: http://your_ip:8080
• Neo4j Browser: http://your_ip:7474
• API: http://your_ip:8000