>AI Security Tools

Galah: An LLM-powered web honeypot.

HexStrike AI MCP Agents is an advanced MCP server that lets AI agents (Claude, GPT, Copilot, etc.) autonomously run 150+ cybersecurity tools for automated pentesting, vulnerability discovery, bug bounty automation, and security research. Seamlessly bridge LLMs with real-world offensive security capabilities.

SCAM - Security Comprehension Awareness Measure | Open-source benchmark that tests AI agents' security awareness during realistic, multi-turn workplace tasks.

MCP Snitch is a macOS application that intercepts and monitors MCP server communications, providing security analysis, access control, and audit logging for AI tool usage.

A tool for folks who `git clone` first and ask questions later

Cybersecurity AI (CAI), the framework for AI Security

A security system to protect your vibecoded apps

An agent that can help triage vulnerabilities across multiple codebases

An AI-powered security review GitHub Action using Claude to analyze code changes for security vulnerabilities.

Open-source EDR for AI agents. Monitor processes, files, network, and behavior of autonomous AI agents.

A collection of servers which are deliberately vulnerable to learn Pentesting MCP Servers.

A Burp Suite extension that integrates OpenAI's GPT to perform an additional passive scan for discovering highly bespoke vulnerabilities and enables running traffic-based analysis of any type.

Metis is an open-source, AI-driven tool for deep security code review

This script automates SQL injection testing using SQLMap with AI-powered decision making.

Claude MCP server to perform analysis on ROADrecon data

a CLI that provides a generic automation layer for assessing the security of ML models

AI EDR for developer workstations and autonomous agent fleets. Build Swarm Detection & Response platforms with Clawdstrike.

Automated prompt-based testing and evaluation of Gen AI applications

A secure low code deception runtime framework, leveraging AI for System Virtualization.

A Python toolbox to create adversarial examples that fool neural networks in PyTorch, TensorFlow, and JAX

Use LLMs for document ranking

poutine, a supply chain vulnerability scanner for build pipelines

A CI/CD Red Team Framework for demonstrating Build Pipeline security risks.

Audit and log all GitHub Copilot MCP tool calls in VSCode with ease.

HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal

Scan A2A agents for potential threats and security issues

Scan MCP servers for potential threats & security findings.

Security Scanner for Agent Skills

An adversarial example library for constructing attacks, building defenses, and benchmarking both

DeepTeam is a framework to red team LLMs and AI agents.

A powerful tool for automated LLM fuzzing. It is designed to help developers and security researchers identify and mitigate potential jailbreaks in their LLM APIs.

⚡ Vigil ⚡ Detect prompt injections, jailbreaks, and other potentially risky Large Language Model (LLM) inputs

Utilizng an MCP Server to communicate with your C2

Code Repository for: AIRTBench: Measuring Autonomous AI Red Teaming Capabilities in Language Models

An easy-to-use Python framework to generate adversarial jailbreak prompts.

Official Elastic Skills

OBLITERATE THE CHAINS THAT BIND YOU

Manage / Proxy / Secure your MCP Servers

A flexible framework for security teams to build and deploy AI-powered workflows that complement their existing security operations.

A growing collection of MCP servers bringing offensive security tools to AI assistants. Nmap, Ghidra, Nuclei, SQLMap, Hashcat and more.

A prompt-based pipeline for finding, validating, and proving vulnerabilities using LLM sub-agents.

Raptor turns Claude Code into a general-purpose AI offensive/defensive security agent. By using Claude.md and creating rules, sub-agents, and skills, and orchestrating security tool usage, we configure the agent for adversarial thinking, and perform research or attack/defense operations.

🐢 Open-Source Evaluation & Testing library for LLM Agents

Fast and accurate AI powered file content types detection

LLM powered fuzzing via OSS-Fuzz.

AI-based implant feature

enject: Hide .env secrets from prAIng eyes: secrets live in local encrypted stores (per project) and are injected directly into apps at runtime, never touching disk as plaintext.

Automated Penetration Testing Agentic Framework Powered by Large Language Models

A Model Context Protocol server that connects AI assistants like Claude to AWS security services, allowing them to autonomously query, inspect, and analyze AWS infrastructure for security issues and misconfigurations.

Adding guardrails to large language models.

Damn Vulnerable MCP Server

Security audit tool for Claude Desktop and Claude Code on macOS — single-command visibility into MCP servers, extensions, plugins, connectors, scheduled tasks, and permissions.

Tracking vulnerabilities contributed by AI-written code

The best way to get AI coding agents to solve hard problems in complex codebases.

Diffprivlib: The IBM Differential Privacy Library

Hybrid Autonomous Logic Operator

MCP Bridge is a lightweight, fast, and LLM-agnostic proxy for connecting to multiple Model Context Protocol (MCP) servers through a unified REST API. It enables secure tool execution across diverse environments like mobile, web, and edge devices. Designed for flexibility, scalability, and easy integration with any LLM backend.

Code snippets to reproduce MCP tool poisoning attacks.

Helping Ethical Hackers use LLMs in 50 Lines of Code or less..

JailbreakBench: An Open Robustness Benchmark for Jailbreaking Language Models [NeurIPS 2024 Datasets and Benchmarks Track]

A comprehensive security scanner for Model Context Protocol (MCP) servers that detects vulnerabilities and security issues in your MCP server implementations.

Security preflight and guardrails for OpenClaw/Moltbot

Shannon is an AI pentester for web applications and APIs. It analyzes your source code, identifies attack vectors, and executes real exploits to prove vulnerabilities before they reach production.

Detect shadow AI agents by monitoring network traffic and classifying process-to-domain pairs.

OpenAnt from Knostic is the leading open source LLM-based vulnerability discovery product, helping defenders proactively find verified security flaws while minimizing both false positives and false negatives. Stage 1 detects. Stage 2 attacks. What survives is real.

Detection scripts for MDM deployment to identify OpenClaw installations on managed devices.

Telemetry for OpenClaw - Captures tool calls, LLM usage, agent lifecycle, and message events. Outputs to JSONL file and optionally to syslog for SIEM integration.

AI-driven vulnerability discovery and live validation

A plugin-based gateway that orchestrates other MCPs and allows developers to build upon it enterprise-grade agents.

MCP Server for Ghidra

This repository provides a benchmark for prompt injection attacks and defenses in LLMs

SecureMCP is a security auditing tool designed to detect vulnerabilities and misconfigurations in applications using the [Model Context Protocol (MCP)](https://modelcontextprotocol.io/introduction). It proactively identifies threats like OAuth token leakage, prompt injection vulnerabilities, rogue MCP servers, and tool poisoning attacks.

This repository contains Cursor Security Rules designed to improve the security of both development workflows and AI agent usage within the Cursor environment. These rules aim to enforce safe coding practices, control sensitive operations, and reduce risk in AI-assisted development.

Feed it a number. Your cloned voice does the social engineering, while you sip your coffee. A ghost that talks on the phone for you.

A proof‑of‑concept C2 framework that uses Server‑Sent Events (SSE) and the MCP protocol for agent registration, command dispatch, and result collection. By tunneling through ngrok, you can quickly expose your C2 server to the public internet for rapid testing and demonstration.

Set of tools to assess and improve LLM security.

MCP to help Defenders Detection Engineer Harder and Smarter

AI Red Teaming playground labs to run AI Red Teaming trainings including infrastructure.

Security scanner detecting Python Pickle files performing suspicious actions

🧠 LLMFuzzer - Fuzzing Framework for Large Language Models 🧠 LLMFuzzer is the first open-source fuzzing framework specifically designed for Large Language Models (LLMs), especially for their integrations in applications via LLM APIs. 🚀💥

Visual testing tool for MCP servers

BloodHound-MCP-AI is integration that connects BloodHound with AI through Model Context Protocol, allowing security professionals to analyze Active Directory attack paths using natural language instead of complex Cypher queries.

AI-powered reverse engineering assistant that bridges IDA Pro with language models through MCP.

Agentic LLM Vulnerability Scanner / AI red teaming kit 🧪

IronClaw is an Agent OS focused on privacy, security and extensibility

Sandbox any AI agent in seconds - zero setup, zero latency.

SAST + LLM Interprocedural Context Extractor

NOVA: The Prompt Pattern Matching

Nova-Proximity is a MCP and Agent Skills security scanner powered with NOVA

NOVA - Claude Code Protection System against prompt injection attacks

NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems.

the LLM vulnerability scanner

OpenShell is the safe, private runtime for autonomous AI agents.

The Python Differential Privacy Library. Built on top of: https://github.com/google/differential-privacy

Fully automatic censorship removal for language models

Hands-on MCP security lab: 10 real incidents reproduced with vulnerable/secure MCP servers, pytest regressions, and Claude/Cursor battle-tested exploit walkthroughs

A complete security skill suite for OpenClaw, Hermes, PicoClaw and NanoClaw agents (and variants). Protect your SOUL.md (etc') with drift detection, live security recommendations, automated audits, and skill integrity verification. All from one installable suite.

Make your GenAI Apps Safe & Secure :rocket: Test & harden your system prompt

Test your prompts, agents, and RAGs. Red teaming/pentesting/vulnerability scanning for AI. Compare performance of GPT, Claude, Gemini, DeepSeek, and more. Simple declarative configs with command line and CI/CD integration. Used by OpenAI and Anthropic.

A collection of real world AI/ML exploits for responsibly disclosed vulnerabilities

Protection against Model Serialization Attacks

Zero shot vulnerability discovery using LLMs

A secure* runtime for autonomous AI agents. Policy from plain-English constitutions. (*https://ironcurtain.dev)

An autonomous LLM-agent for large-scale, repository-level code auditing

TextAttack 🐙 is a Python framework for adversarial attacks, data augmentation, and model training in NLP https://textattack.readthedocs.io/en/master/

A CLI tool for threat modeling and visualizing AI agents built using popular frameworks like LangGraph, AutoGen, CrewAI, and more.

Simple Prompt Injection Kit for Evaluation and Exploitation

Security scanner for MCP servers

Open-source AI coworker, with memory

Model Context Protocol (MCP) server to pin components to their immutable version

An AI-powered agentic red team framework that automates offensive security operations, from reconnaissance to exploitation to post-exploitation, with zero human intervention.

Automatic Exploit Generation with LLMs

Semgrep Pro Rules to ensure code using LLMs is following best practices

Jagged Frontier: LLM vulnerability detection benchmark harnesses (API + Claude Code agentic)

Security scanner for AI agents, MCP servers and agent skills.

GitHub Action to alert on security patches before the CVE drops.

A security scanner for your LLM agentic workflows

Scan your dev machine for AI agents, MCP servers, IDE extensions, and suspicious packages - in seconds.

CyberGym is a large-scale, high-quality cybersecurity evaluation framework designed to rigorously assess the capabilities of AI agents on real-world vulnerability analysis tasks.

LLM-based automated patch diffing

An AI-powered tool for discovering privilege escalation opportunities in AWS IAM configurations.

Opinionated defaults, documentation, and workflows for Claude Code at Trail of Bits

Sandboxed devcontainer for running Claude Code in bypass mode safely. Built for security audits and untrusted code review.

A Python pickling decompiler and static analyzer

MCP security wrapper

Trail of Bits Claude Code skills for security research, vulnerability detection, and audit workflows

Adversarial Robustness Toolbox (ART) - Python Library for Machine Learning Security - Evasion, Poisoning, Extraction, Inference - Red and Blue Teams

Open-source AI penetration testing tool to find and fix your app’s vulnerabilities.

a security scanner for custom LLM applications

Vulnerability detection framework by Binarly's REsearch team

Security gateway for AI agents - credential-isolated API proxying and policy-gated remote execution (conclaves). Reduce the blast radius!

A minimal LLM-powered zero-day vulnerability scanner by AISLE.

🔍 LangKit: An open-source toolkit for monitoring Large Language Models (LLMs). 📚 Extracts signals from prompts & responses, ensuring safety & security. 🛡️ Features include text quality, relevance metrics, & sentiment analysis. 📊 A comprehensive tool for LLM observability. 👀

Demos for Black Hat Europe 2025's The Forensic Trail On GitHub: Hunting For Supply Chain Activity

Baseline rules files to improve the security of AI-generated code (Claude, Cursor, Copilot + more)

A simple POC to expose Mythic as a MCP server

LLM Agent Skill for YARA rule authoring and review