#1Adversarial ML attack frameworks, evasion techniques, and model robustness evaluation tools.
Kitploit recommended
hannahxchen/llm-censorship-steeringCode for 'Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control'
marcgoam/cve-2026-54121-certighostCVE-2026-54121 (Certighost) AD CS DC-impersonation PoC. Patched SAN handling + MAQ-safe account reuse.
sonnelon/cve-2026-23744-pocThe poc of CVE-2026-23744
george0papasotiriou/cve-2026-1337-ai-coding-assistant-prompt-injection-to-sandbox-escapeA prompt injection in a code‑review bot that executes AI‑generated fixes in a sandbox. The sandbox uses a blacklist to prevent dangerous commands,…
helixo32/nimblackoutKill AV/EDR leveraging BYOVD attack
idov31/jormungandrJormungandr is a kernel implementation of a COFF loader, allowing kernel developers to load and execute their COFFs in the kernel.
winscripting/uac-bypassPowerShell proof-of-concept that bypasses Windows User Account Control (UAC) to elevate privileges, intended for authorized penetration testing and…
sh0ckfr/lockbit3.0-mpclient-defender-pocLockbit3.0 Microsoft Defender MpClient.dll DLL Hijacking PoC
ioactive/i-know-where-your-page-livesI Know Where Your Page Lives: Derandomizing the latest Windows 10 Kernel - ZeroNights 2016
kudaes/bin-finderDetect EDR's exceptions by inspecting processes' loaded modules
jwillyamz/ezemuSee adversary, do adversary: Simple execution of commands for defensive tuning/research (now with more ELF on the shelf)
choisg/onedriveupdatersideloadingPayload for DLL sideloading of the OneDriveUpdater.exe, based on the PaloAltoNetwork Unit42's blog post
crisprss/magicnetdefsSimilar to Petitpotam, the netdfs service is enabled in Windows Server and AD environments, and the abused RPC method allows privileged processes to…
jackpurcell/autoran-publicAutomated framework for hijacking safety reasoning in large reasoning models via simulated reasoning traces and iterative prompt refinement to…
andyshaw01/poisoncraftThis repository provides the official implementation of POISONCRAFT: Practical Poisoning of Retrieval-Augmented Generation for Large Language Models.
sbamit/exponentiated-gradient-descent-llm-attackA novel adversarial attack on LLM based on the Exponentiated Gradient Descent technique.
george0papasotiriou/cve-2026-22007-ntp-monlist-amplification-over-ipv6Python PoC for CVE-2026-22007: NTP monlist amplification over IPv6, including a simulated vulnerable server and spoofed UDP reflection attack.
