RESK Security Projects

ReskCrypt

Conversation Encryption & Model Monitoring

End-to-end encryption for AI conversations with advanced monitoring capabilities. Features AES-256-GCM encryption, ECDSA digital signatures, and real-time dashboard monitoring.

Key Features

• State-of-the-art encryption (AES-256-GCM, ECDSA)
• Real-time monitoring dashboard
• Secure conversation storage
• Advanced security features

GNOM

Graph Neural Network Monitoring

Advanced security solution powered by Graph Neural Networks (GNN) that transforms network infrastructure into dynamic graph structures for real-time threat detection.

Key Features

• Real-time network topology mapping
• GNN-based threat detection
• Reinforcement Learning attack prediction
• Vulnerability assessment through graph analysis

ReskLayer

Advanced Prompt Injection Detection

Cutting-edge solution for detecting malicious prompt injections using ModernBERT with DiffTransformer attention mechanisms and Ettin's three-phase training recipe.

Key Features

• ModernBERT with DiffTransformer attention
• Detection of DAN and complex attacks
• CUDA optimization and DeepSpeed integration
• PINT testing framework validation

resk-llm

A robust Python library designed to enhance security and manage context when interacting with OpenAI's language models.

resk-mcp

An open-source Python library that adds a robust security and management layer over the official Model Context Protocol (MCP) Python SDK.

resk-llm-ts

A comprehensive security toolkit for JavaScript applications using Large Language Models with built-in security features.