#1Security papers, research projects, technical references, and reproducible material for defensive or authorized analysis.
Kitploit recommended

GhostLock stripped to one primitive: SELinux off on Galaxy A17 (BZA5) via futex PI UAF (CVE-2026-43499). No root, no cred patch, no rwforge.

An Evaluation Agent for Detecting Misinformation and Knowledge Poisoning in Retrieval-Augmented Generation Systems.

Awesome-Cellular-Hacking

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…

Embed multiple secret messages in LLM chat token choices using arithmetic/Discop steganographic coders, with bit-exact decoding and steganalysis…

Vulnerability research write-up on CVE-2017-7921 — a critical unauthenticated auth bypass in Hikvision IP cameras/DVRs/NVRs, covering root cause,…

Research toolkit that roots a Samsung Tizen 9 TV via a Chromium WasmGC RCE and V4L2 physmap arbitrary physical memory access, with persistent on-boot…

Provably secure linguistic steganography embedding secret messages into LLM-generated text via rotation range-coding. Includes embed, extract, and…

[🎖️ Honorable Mention in HAIPS @COLM'26] The first benchmark to evaluate how well do LLM agents act in alignment with their assigned authorization…

Benchmarking prompt injection detections for web agents.

Private end-to-end sanitizer reproduction package for six GDCM findings

Agent Control Protocol (ACP) — Official English specification. Cryptographically verifiable authorization architecture for autonomous AI agents.

Syntactic Ghost: An Imperceptible General-purpose Backdoor Attacks on Pre-trained Language Models

Techniques and tools for Windows kernel exploitation, covering pool overflow exploitation, segment heap metadata abuse, and debugging scripts for…

Detects LLM context-leakage attacks by training lightweight behavior probes on log-probabilities, with vLLM offline/server detection pipelines.

😎 🔗 Awesome list about all kinds of resources for learning Ethical Hacking and Penetration Testing.

GitLab Duo DevSecOps loop where a second AI re-attacks every AI-written fix, saving each successful bypass as a permanent antibody replayed in CI and…

The research UEFI hypervisor that supports booting an operating system.