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

Static security scanner for AI agent skill packages. Detects malicious SKILL.md files and bundled scripts before they run.

The code for ACM MM2024 (Multimodal Unlearnable Examples: Protecting Data against Multimodal Contrastive Learning)

Next-gen logical WAF engine built in SWI-Prolog. Features an inductive learning brain running at 2M+ LIPS with an integrated recursive decoder to…

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.

Code for 'Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control'

Stop prompt injection attacks before they reach your LLM — zero API costs, runs entirely locally, integrates in 2 minutes. Prompt injection is the…

Beyond the Prompt: Jailbreaking Function-Calling LLMs via Simulated Moderation Traces

Scans PostgreSQL databases and Jira issues for PII using regex, NER/ML, and optional self-hosted LLM detection, with redacted reports and audit…

Reference implementation for "Hard Stop: Kernel-Level Preemption and Containment for Rogue Agentic Execution". Out-of-band Epistemic Andon Cord,…

Policy-governed LLMSecOps framework providing AST-based SAST, secret scanning, supply-chain and multi-cloud CSPM checks, AI-BoM generation, and CI/CD…

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

Curated reading list and taxonomy of attack and defense research for mobile on-device AI systems, covering adversarial, backdoor, model stealing, and…

Runtime guard skill that detects and blocks malicious skill-induced attacks in skill-augmented agents, with an MCTS loop that evolves the guard…

Red-teaming benchmark and black-box attack framework for evaluating visual misinformation risks in commercial image generators, with 10k false-claim…

All-source intelligence fusion dashboard — WiFi, cellular, CCTV, ADS-B, orbital & SDR feeds unified into a single tactical map. Security research…

Causal context attribution and rule-based monitor LLM defense against indirect prompt injection in LLM agents, achieving state-of-art performance on…

MCP server that lets AI assistants drive a self-hosted Sn1per install, running scoped, audited scans and returning findings, ports, subdomains and…

Host-agnostic pre-write security hook for coding agent: detects user-input patterns via Semgrep and emits deterministic, no-LLM security guidance.