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

OWASP project focused on security for large language model applications, covering LLM-specific risks, testing guidance, and mitigation practices.

OpenAI compaction research: GhostCompact + Compaction Smuggling

Research codebase for privacy evaluation of quantized neural networks: trains target/shadow models under PTQ/QAT and runs membership-inference…

Benchmark code and interactive results for autonomous network-attack detection by tool-using LLM agents investigating raw PCAPs.

Fork optimizado de Bytebot: Agente de IA para control de escritorio. Reducido de 7GB a 1.09GB, parcheada CVE-2026-78250, con documentación viva…

Proof-of-concept and full writeup for CVE-2026-47644: Edge Copilot prompt injection via attachment tag breakout leading to cross-tab data…

A complete guide and workflow for integrating Agile sprints with DevOps CI/CD pipelines.

Express/TypeScript microservice with a Gemini-powered dual-agent consensus loop that synthesizes and validates advisory content for CVE-2026-96230 in…

Fire ChatML/Mistral/Llama control-token literals at your self-hosted agent stack and prove role-forgery before an attacker does (CVE-2026-61732).

A honeypot that disrupts AI-agent-driven attacks with tarpitting and "context bomb" techniques. Python/Flask, config-driven, with multi-channel…

Claude Code skill: read-only check whether your ComfyUI is affected by CVE-2026-123790 (Easy-Use path traversal to RCE): version, fix commit,…

Public vulnerability advisories with root-cause write-ups and reproducible PoCs for coordinated-disclosure findings in Psono Server and Microsoft…

Generation-time safety framework for LLMs that maps internal model signals to explicit rules and per-request policies, enforcing allow, refuse, or…

Training-free, trigger-agnostic test-time detector for backdoored Vision Transformers that flags poisoned samples using inter-head attention…

Taxonomy-guided adaptive red teaming framework that sweeps prompt-injection techniques and refines failed attempts to find diverse attacks against…

Scans WordPress plugins for missing-guard vulnerabilities using ripgrep signatures, Semgrep rules, attack-surface mapping, and an LLM review cascade…

Research pipeline implementing a visual-contextual jailbreak against multimodal LLMs: T2I prompt extraction, SD3.5 image rendering, multi-round spoof…

Benchmarking Bias in LLM Agents for Cyber-Attack Scenarios