Fastest Claws in the West

🦀 Network Security Monitor

🎯 Project Overview

Demo Screenshot

#rust #security #privacy #ai #crossplatform #mongodb #streamlit #llm

A privacy-first network security monitoring tool that analyzes your local network traffic for potential security threats. Built with Rust for performance and reliability, featuring AI-powered threat detection while keeping your data private.

Key Features

🔒 Privacy-First Design: Only analyzes metadata, not packet contents
🚀 High-Performance: Written in Rust for blazing-fast packet processing
🤖 AI-Powered Insights: LLM-based threat analysis without exposing sensitive data
📊 Real-Time Dashboard: Built with Streamlit and async MongoDB integration
🌐 Cross-Platform: Works on Linux, macOS, and Windows
💻 On-Device Processing: All analysis happens locally except for anonymized LLM queries

Component Diagram

Architecture Diagram

🏗️ Technical Achievements

Rust Implementation: Learned and implemented core networking functionality in Rust
Async Streamlit: Successfully integrated async MongoDB operations with Streamlit
Privacy-Preserving Design: Developed a secure architecture that respects user privacy
Cross-Platform Networking: Handled platform-specific network capture requirements
Real-Time Processing: Efficient packet processing and analysis pipeline

🚀 Getting Started

Quick Start

Unix/MacOS

# Clone the repository
git clone https://github.com/nichoc0/ConUHacks2025

Windows

# Clone the repository
git clone https://github.com/nichoc0/ConUHacks2025

Manual Setup

If you prefer to set up manually:

# Install Rust dependencies
cargo build

# Install Python dependencies
pip install -r requirements.txt

# Start MongoDB (required for event storage)
mongod

# Run the Rust backend
cargo run

# Launch the dashboard
streamlit run streamlit_dashboard.py

🔒 Legal & Privacy

This tool is designed for monitoring your own network traffic only. By default, it operates in non-promiscuous mode, capturing only your machine's traffic to ensure legal compliance. The promiscuous mode toggle should only be used in environments where you have explicit permission to monitor network traffic.

Privacy Features

Only metadata is analyzed, never packet contents
All processing happens on your device
LLM queries are anonymized and contain no identifying information
Local MongoDB storage with configurable retention

🛠️ Technical Architecture

Core Components

Packet Capture: Rust-based packet sniffing
Threat Detection: - Pattern recognition of common threats
Event Storage: Local MongoDB with async operations
Dashboard: Streamlit interface with real-time updates

💡 Challenges & Learning

Rust Learning Curve: Overcame not knowing any Rust
Async Integration: Successfully overcame Streamlit's flawed async handling with async MongoDB
Privacy Design: Balanced security insights with data privacy