
PurpleLlama
Set of tools to assess and improve LLM security.

Set of tools to assess and improve LLM security.

The AI Security Verification Standard (AISVS) focuses on providing developers, architects, and security professionals with a structured checklist to…

Anti-LLM obfuscation via finger counting

A prompt injection in a code‑review bot that executes AI‑generated fixes in a sandbox. The sandbox uses a blacklist to prevent dangerous commands,…

Two-stage prompt-injection and jailbreak detector: regex gates plus a quantised DeBERTa-v3 ONNX classifier, with image, document, and audio support.…

The Security Toolkit for LLM Interactions

Scalable assembly analysis platform for indexing, clone search, and executable classification using static, dynamic, and machine-learning techniques…

Code for ACL 2026 (main) paper "DeepGuard: Secure Code Generation via Multi-Layer Semantic Aggregation"