
llm-security-101
Delving into the Realm of LLM Security: An Exploration of Offensive and Defensive Tools, Unveiling Their Present Capabilities.

Delving into the Realm of LLM Security: An Exploration of Offensive and Defensive Tools, Unveiling Their Present Capabilities.

Security scanner for AI/ML model files. Detects malicious code, backdoors, and vulnerabilities before deployment

Windows 7 UAC Bypass Vulnerability in the Windows Script Host

Proof-of-concept exploit that bypasses ASLR on Intel CPUs by abusing branch target buffer and speculative execution to leak randomized addresses via…

Benchmarking prompt injection detections for web agents.

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

CEREBRO-RED v2: Advanced LLM Red Team Research Platform with PAIR Algorithm and LLM-as-a-Judge Evaluation

CVE-2026-54121 (Certighost) AD CS DC-impersonation PoC. Patched SAN handling + MAQ-safe account reuse.

This repository provides the official implementation of POISONCRAFT: Practical Poisoning of Retrieval-Augmented Generation for Large Language Models.

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,…

Defense framework that keeps audio-language models frozen and adds a mid-layer risk gate with late-layer safety adapters to block audio jailbreaks at…

Research code for a gray-box trojan attack that flips a single KV-cache bit in fine-tuned LLM classifiers and measures per-class attack success rate.

Experiments for control-token chain-of-thought suppression and parser-leniency attacks on tool-using LLM agents

Black-box input-stage purification defense that neutralizes backdoor attacks on object detectors via corruption, diffusion reconstruction, and DBSCAN…

XSS2Shell (CVE-2026-64638) WordPress pre-auth XSS to RCE PoC mirror — WordSec, MIT; for authorized security testing

Exploit PoCs for CVE-2025-30374, a Taipy class pollution bug, demonstrating RCE, reflected XSS, DoS, and OpenAI credential leakage with Docker-based…

CVE-2026-6765, Test only FormAutofill handlers exposed in Firefox

CVE-2026-7070 PoC for RDP clipboard hijacking via virtual channel injection; includes simulated server and exploit script for data theft/credential…