
robustbench
Standardized adversarial robustness benchmark with a public leaderboard and downloadable model zoo for evaluating ML models against Lp attacks and…

Standardized adversarial robustness benchmark with a public leaderboard and downloadable model zoo for evaluating ML models against Lp attacks and…

Test your prompts, agents, and RAGs. Red teaming/pentesting/vulnerability scanning for AI. Compare performance of GPT, Claude, Gemini, DeepSeek, and…

Adversary simulation and Red teaming platform with AI

Automated behavioral evaluation framework for LLMs that generates diverse test scenarios to probe for sycophancy, bias, and other safety-relevant…

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

Bypassing UAC with SSPI Datagram Contexts

Train, evaluate, and explore neural networks with built-in adversarial robustness tools, including PGD attacks, adversarial training, and robust…

PurpleSharp is a C# adversary simulation tool that executes adversary techniques with the purpose of generating attack telemetry in monitored Windows…

Playing around with Stratus Red Team (Cloud Attack simulation tool) and SumoLogic

HyperDeceit is the ultimate all-in-one library that emulates Hyper-V for Windows, giving you the ability to intercept and manipulate operating system…

Weaponizing to get NT SYSTEM for Privileged Directory Creation Bugs with Windows Error Reporting

A PoC implementation for dynamically masking call stacks with timers.

Encypting the Heap while sleeping by hooking and modifying Sleep with our own sleep that encrypts the heap

PE obfuscator with Evasion in mind

Crystal port of GodPotato to abuse SeImpersonatePrivilege with indirect syscalls, dynamic API resolution and compile-time string obfuscation. Run…

See adversary, do adversary: Simple execution of commands for defensive tuning/research (now with more ELF on the shelf)

PoC repository for the blog post CopyEscape: Taking Over Docker Hosts with docker cp

A high-severity prompt injection flaw in Claude AI proves that even the smartest language models can be turned into weapons — all with a few lines of…