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

An alignment auditing agent capable of quickly exploring alignment hypothesis

HopLa Burp Suite Extender plugin - Brings AI capabilities, autocompletion support, and a set of useful payloads to Burp Suite

ExploitGym is a large-scale, realistic benchmark built from real-world vulnerabilities designed to evaluate AI agents' ability to develop exploits.

AI-powered vulnerability scanner extension for Burp Suite with multi-provider support (Ollama, OpenAI, Claude, Gemini)

AI-Assisted Reverse Engineering with Ghidra

opensource repo for validating agentic AI applications: redteam, behavior, supply-chain, static analysis

Multi-Source Analysis & Threat Intelligence Search

Fine-tuning framework that applies first-order optimal safety calibration and periodic recalibration to LLMs, preserving safety-compatible updates…

Embeds and detects keyed, style-based watermarks in LLM-generated Python, Java, and C++ code via CST rewriting, preserving functional correctness and…

Open-source antivirus and firewall for AI agents: block risky tools, secret access, prompt injection, malicious packages, MCP servers, plugins, and…

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

Burp Suite extension that adds built-in MCP tooling, AI-assisted analysis, privacy controls, passive and active scanning and more

Kernel-level eBPF sandbox for securing LLM agent tool calls made through the Model Context Protocol (MCP)

Token by Token, Compromised: Backdoor Vulnerabilities in Unified Autoregressive Models (NeurIPS'26)

Research implementation of Hop-Decayed Influence (HDI) and the 3S attack framework, exposing structural auxiliary indexing vulnerabilities in…

LLM safety: analyze how fine-tuning shifts refusal behavior and evaluate layer-freezing and low-rank defenses.

An open, local-first security testing platform for pentesters, AI agents, CI/CD pipelines, and teams.

An empirical security testbed evaluating prompt injection, confused-deputy vulnerabilities, and tool-calling defenses in LLM agents.