#1LLM security, prompt injection, model extraction, adversarial AI, and AI red teaming tools.
Kitploit recommended

OpenMed < 1.5.2 unauthenticated RCE via PII privacy-filter model loading and trust_remote_code=True

Self-referenced local contrast for knowledge-poison detection in retrieval-augmented generation

A privacy-first app that strips AI watermarks from content you own.

Cybersecurity writeups, walkthroughs, and technical notes by Nanashi Bx2.

A curated collection of resources for learning and researching LLM prompt injection attacks, defenses, and security.

AISA-Scanner is an AI-powered autonomous vulnerability scanner that maps CVEs to metasploit exploits, MITRE, CEH, and SANS, delivering intelligent,…

This repository contains the complete record of my three-year research journey, covering the project from foundational concepts to advanced-level…

Ghostsplice repository: PoC for Cross-Channel Trust Fragmentation Attack

Self-hostable AI SOC that fuses security alerts, auto-triages via agentic AI, runs MITRE ATT&CK investigations, and logs every agent decision in a…

A font-based deception tool for red teaming, security research, and whatever else.

Runtime security gateway for AI agents: cryptographically attests tool calls, enforces policies, sandboxes execution, and logs tamper-evident audit…

Docker Container Escape POC via mlx-metal importlib

Docker Model Runner container-to-host RCE / Escape: A critical vulnerability that allows for container-to-host code execution in the Docker Model…

CVE-2026-20685 - Draft or TODO

Benchmarking framework for evaluating computer-use AI agents against multi-step indirect prompt injection, with automatic adversarial goal…

Analyzes LLM internal states and 100+ attention/probability features to train classifiers that detect document poisoning attacks in RAG systems.

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

Redefining vulnerability disclosure in the AI era. We mass-produce exploitable 0days and disclose them directly, using event-driven pressure to…