Security & AI safety

Pentesting & offensive security

763 agents · most starred first

Pentesting & offensive security, page 4 of 8

zebbern-kali-mcp★ 54

MCP server for Kali toolkit (docker) - 121 tools for AI-assisted security testing give agents access to all it needs!

witchcraft_fex★ 54

WitchcraftFEX Are much more than a simple hacking script, is a cyberdeck toolkit built for runners who dive deep into the net. A shell and automation's and automation's engine for pentesting, osint, forensics, fingerprinting,web-hacking, malware building, reverse shells, RCE agents, generation of reports (RBR), and playbooks (RBP) based on rules.

cybermind★ 53

⚡ AI-powered cybersecurity CLI tool for Kali Linux | Ethical hacking | Penetration testing | Bug bounty

WonderSuite-Ai-Bug-Bounty★ 52

AI-Powered Offensive Security Research Engine - desktop-native security testing platform with native MCP integration. 90 tools, MITM proxy, stealth browser, autonomous AI agent. Built on Tauri + Rust + React.

LLMFuzzAgent★ 51

[Corca / ML] Automatically solved Gandalf AI with LLM

FB-hack★ 51

#decompiled by PDM31 import os, sys print '\x1b[1;32mSudah punya ID dan Password nya?' print '\x1b[1;32mSilahkan Login ' import os, sys def wa(): os.system('xdg-open https://api.whatsapp.com/send?phone=6281291977644&text=Assalamualaikum') def restart(): ngulang = sys.executable os.execl(ngulang, ngulang, *sys.argv) user = raw_input('ID: ') import getpass sandi = raw_input('Password: ') if sandi == 'indoxploit' and user == 'Borot': print 'Anda Telah Login' sys.exit else: print 'Login GAGAL, Silahkan hubungi ADMIN' wa() restart() import os, sys, time, datetime, random, hashlib, re, threading, json, getpass, urllib from multiprocessing.pool import ThreadPool try: import mechanize except ImportError: os.system('pip2 install mechanize') else: try: import requests except ImportError: os.system('pip2 install requests') from requests.exceptions import ConnectionError from mechanize import Browser reload(sys) sys.setdefaultencoding('utf8') br = mechanize.Browser() br.set_handle_robots(False) br.set_handle_refresh(mechanize._http.HTTPRefreshProcessor(), max_time=1) br.addheaders = [('User-Agent', 'Opera/9.80 (Android; Opera Mini/32.0.2254/85. U; id) Presto/2.12.423 Version/12.16')] def keluar(): print '\x1b[1;91m[!] Keluar' os.sys.exit() def jalan(z): for e in z + '\n': sys.stdout.write(e) sys.stdout.flush() time.sleep(0.1) logo = '\x1b[1;92m\n\xe2\x95\x94\xe2\x95\xa6\xe2\x95\x97\xe2\x94\x8c\xe2\x94\x80\xe2\x94\x90\xe2\x94\xac\xe2\x94\x80\xe2\x94\x90\xe2\x94\xac\xe2\x94\x8c\xe2\x94\x80 \xe2\x95\x94\xe2\x95\x90\xe2\x95\x97\xe2\x95\x94\xe2\x95\x97 \n \xe2\x95\x91\xe2\x95\x91\xe2\x94\x9c\xe2\x94\x80\xe2\x94\xa4\xe2\x94\x9c\xe2\x94\xac\xe2\x94\x98\xe2\x94\x9c\xe2\x94\xb4\xe2\x94\x90\xe2\x94\x80\xe2\x94\x80\xe2\x94\x80\xe2\x95\xa0\xe2\x95\xa3 \xe2\x95\xa0\xe2\x95\xa9\xe2\x95\x97\n\xe2\x95\x90\xe2\x95\xa9\xe2\x95\x9d\xe2\x94\xb4 \xe2\x94\xb4\xe2\x94\xb4\xe2\x94\x94\xe2\x

stratum-c2★ 51

Cloud dead-drop C2 framework — RSA-4096 + AES-256-GCM, 5 cloud providers, Rust-only agents, P2P mesh, persistence engine, credential harvesting

Sphinx.bot★ 50

Sphinx.bot is a platform for collecting data leaks. All information recorded here is publicly available on the web. Any malicious use of this information is strictly prohibited.

Cre4T3Tiv3★ 50
faraday_agent_dispatcher★ 50

Faraday Agent Dispatcher launches any security tools and send results to Faradaysec Platform.

forge★ 50

Command Augmentation support for BOFs and .NET assemblies across agents

sage★ 50

Sage is a virtual Mythic agent that that uses an AI agentic system to operate Mythic and Mythic agents running on compromised hosts.

bugbounty-agent★ 50

Automated bug bounty reconnaissance and scanning agent

nua-bench★ 50

Open-source benchmark of 5 intentionally vulnerable web apps with 100 baseline vulnerabilities for evaluating AI penetration testing agents.

DVLAA★ 50

DVLAA (Damn Vulnerable LLM and Agent Application) is a local LLM and Agent security range for OWASP LLM Top 10 and Agent security training.

DVAIA-Damn-Vulnerable-AI-Application★ 50

Damn Vulnerable AI Application - For LLM Red Team Training. LLM testing, RAG testing, Multimodal testing, Agent testing, LLM paload generation

Hackbot-Arena★ 50

⚔️ Benchmark your AI agent with 30 Docker-based web security labs inspired by real bug bounty findings, complete with canonical flags and reference solvers.

AuthKit★ 50

首个多维度,UI/UX友好的Burp suite越权漏洞检测插件

era-agent★ 50

ERA – Evolve & Reverse Engineer Anything: an AI-powered reverse-engineering agent built on REA, with a controlled, evidence-driven Improvement Lab. Reverse. Analyze. Evolve. Repeat.

Sliver-CPPImplant2★ 49

Sliver agent rewritten in C++

sliver★ 49

Sliver agents for Mythic

worm-gpt★ 49

WormGPT CLI: Chaos engine for testing LLM workflows.

mtk-payloads★ 49

Payloads for MediaTek Download Agents

security-research-orchestrator-prompt★ 49

High-assurance, artifact-agnostic prompt for authorized security labs, exploit-chain research, PoC validation, and evidence-led reporting.

src-6k-security-research-skills★ 49

SRC-6K Security Research Skills: black-box SRC and white-box audit agent package

worm-gpt★ 49

WormGPT CLI: Chaos engine for testing LLM workflows.

vigilo★ 49

AI-powered penetration-testing pipeline. Specialised agents recon, exploit and adversarially verify web-application vulnerabilities, then produce evidence-backed reports and per-finding fix branches. White-box (source) and grey-box (no-source) modes. Built by CERT-EU.

HavocHub★ 48

PoC for a Havoc agent/handler setup with all C2 traffic routed through GitHub. No direct connections: all commands and responses are relayed through Issues and Comments for maximum stealth.

Awesome-Hacking-with-AI★ 48

A curated, evidence-led guide to AI for security and security for AI. Defensive research, agent safety, MCP security, AI red-teaming and evaluation, authorized testing workflows, and detection engineering.

paranoid★ 48

Your app is guilty until proven secure: an agent skill whose /hack-me breaks into your own running app, proves each bug with a real request, patches it, and re-verifies. Claude Code · Codex · Cursor.

macta★ 47

MACTA: A Multi-agent Reinforcement Learning Approach for Cache Timing Attacks and Detection

CVE-2022-46169★ 47

CVE-2022-46169 Cacti remote_agent.php Unauthenticated Command Injection.

CTF-Dojo★ 47

Training Language Model Agents to Find Vulnerabilities with CTF-Dojo

kali-linux-image★ 47

Lightweight Kali Linux Docker images with curated CLI penetration testing tools for headless environments

offense★ 47

OpenRouter for agent security.

inspect_cyber★ 47

An Inspect extension for agentic cyber evaluations

CrystalForge★ 46

AdaptixC2 default beacon agent extended to support Crystal Palace loaders.

1337GPT★ 46

My attempt at making a GPT agent for pentesting

nyuctf_agents_craken★ 46
REvengeBS★ 46

Frida agent for Brawl Stars v68.250 (official, arm64). Aimbot, ESP, killaura, autododge.

ThunderStorm★ 46

Golang C2 Server and Agents using XMT (https://github.com/iDigitalFlame/xmt)

Hephaestus★ 46

Hephaestus 是集资产收集、攻击面评估、漏洞验证与 AI 渗透对话于一体的安全测试平台。提供 Web 服务、桌面客户端及独立扫描 Agent,支持本机扫描、组合任务和分布式节点执行。

benchmark-platform★ 45

AI agent CTF 靶场竞赛平台

IDACLI★ 45

AI-native IDA Pro kernel — unrestricted IDAPython over JSONL, persistent sessions, zero dependencies. Let your agent drive IDA, not call tools.

drift★ 45

Security AI helper for Caido: pipes your local Claude Code / Gemini / Codex / Copilot CLI through 18 MCP tools for manual web security testing. Local-first, no API keys.

needle-agent★ 44

The iOS Agent for the Needle Security Assessment Framework

pentesting-mcp-servers-checklist★ 44

A practical, community-driven checklist for pentesting MCP servers. Covers traffic analysis, tool-call behavior, namespace abuse, auth flows, and remote server risks. Maintained by Appsecco and licensed for remixing.

MobHunt★ 44

An agentic mobile security research system built as a Claude Code project. An orchestrator spawns focused sub-agents, each with one narrow objective.

llm-sec-range★ 44

🛡️ 大模型攻防渗透测试靶场 · 提示注入CTF / OWASP LLM Top10 / 脆弱Agent / 资料聚合。目标模型可切换:DeepSeek直连、OpenRouter中转站(国产+国外十余款小模型)、本地Ollama(DeepSeek-R1 8B离线)。一键本地部署。

Hexestra★ 44

Hexestra is an AI-native penetration testing IDE where human operators and the AI Agent work on the same operational surface—sharing the browser, terminal, traffic, shells, asset graph, task plan, evidence, and controls.

Atlas★ 44

Open-source local AI Agent desktop app for authorized security assessment — Electron + Mastra-powered runtime, visual pentest workflow, Kali sandbox, multi-provider LLM routing. 本地 AI 安全评估智能体桌面应用

Agentic-AI-for-Offensive-Cybersecurity★ 44

Agentic AI for Offensive Cybersecurity, published by Packt

grepleaks★ 44

Grepleaks — your AI pentest environment. An AI pentest agent that lives in your terminal.

metasploit-gym★ 43

An environment for testing AI agents against networks using Metasploit.

hackbot★ 43

Production-ready AI-powered pentesting & cybersecurity assistant with Agent, Chat & Planning modes, native desktop GUI, 10 AI providers, and built-in intelligence modules.

DRX-Operator★ 42
fawkes★ 42

Fawkes is a golang Mythic C2 Agent exclusively written by AI.

ida-swarm★ 41

An AI-powered autonomous reverse engineering swarm that accelerates binary analysis through multi-agent collaboration

TRACE★ 41

TRACE is an agentic jailbreaking framework that can reduce attack difficulty, induce the agent to execute end-to-end attack workflows through task decomposition, execution-oriented multi-turn interaction, and adaptive recovery.

ai-red-teaming★ 41

Whitebox & Blackbox AI red-teaming framework for LLMs & Agentic AI apps. It analyzes your app's source code to discover tools, roles, and guardrails, then generates new attacks chains across several categories and adapts over multiple multi turn rounds to find vulnerabilities

LingXi★ 40

LingXi: multi-agent CTF and pentest framework originated from Tencent Security Hackathon

RuwoScan★ 40

基于多智能体架构与自主决策机制,配合大语言模型的理解能力,输入自然语言指令,AI 自动完成从目标侦察到漏洞验证的完整渗透测试流程。

sapper★ 40

Engagement fact graph for autonomous pentest agents — typed facts, flag ledger, arena adapter, offline KB. Offensive fork of claude-fact-graph.

wraith★ 39

WRAITH — Weakness Reasoning & AI Threat Hunter | Multi-agent RL-driven penetration testing framework | DEF CON 2026

Gideon★ 38

Open-Source autonomous security operations and red teaming agent built to help defenders investigate threats, analyze vulnerabilities, assess indicators of compromise, generate hardening guidance, and execute security research through an auditable agent workflow.

kevlar-benchmark★ 37

Kevlar Benchmark: OWASP Top 10 for Agentic Apps (AI-Agents) 2026 a Red Team Benchmark

CyberChef-MCP★ 37

Model Context Protocol server for CyberChef — exposes GCHQ's "Cyber Swiss Army Knife" as AI-agent tools: 504 operations across encryption, encoding, compression and forensics, on MCP protocol revision 2026-07-28

Null-CLi★ 37

Open-source AI pentest and compliance-readiness CLI by NullSquare.

myrtia★ 37

Myrtia — agent-native red teaming for LLMs and AI agents: a deliberating council of attacker models (propose, cross-critique, Borda vote, chairman synthesis), 13 attack strategies, live turn-by-turn UI and ASR analytics. Open-source, built on IBM ARES.

cyber-safari★ 36

A fun POC that is built to understand AI security agents.

bugbountyrules★ 36

Behavioral-discipline skill turning an AI agent into a methodical, non-robotic bug-bounty & pentest hunter - 42 always-on rules: scope, coverage, no false positives, never quits early.

PhantomPipe★ 35

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.

nikki-ai-cli-assistent★ 35

🫡 Nikki — AI Assistant for Red Team & DevOps in the Terminal (fish-shell)

pro-pentagi-AI-Agents-system★ 35

Fully autonomous AI Agents system capable of performing complex penetration testing tasks

pentest_skill★ 35

Black-box web penetration testing automation framework for AI Agents

codex-red-team-prompt★ 35

OpenAI Codex 自定义系统提示词注入工具 | System Prompt Injection | CTF Security Research Red Team | Auto Install | Cross-Platform

PentestPi★ 35

Autonomous Penetration Testing Agent Framework with External State Memory & Anti-False-Falsification Guardrails.

CheckmateC2★ 34

CheckmateC2 is a custom agent and listener for the Havoc Framework (made by @Cracked5pider) that communicates entirely through Chess.com game collections and analysis boards using Base5 and FEN for data encoding.

red-team-blue-team-agent-fabric★ 34

AI agent security harness, adversarial testing: 661 executable tests in the v4.31.1 release, across MCP, A2A, x402/L402, decision governance, benchmark integrity, human-in-the-loop, skill supply chain. Five target shapes: a verdict must be able to be wrong AND to be right. OWASP Agentic v1.1 T1-T17, AIUC-1 crosswalk (2026-10-15), NIST AI 800-2.

blackbox-re-agent★ 34

Conversational black-box reverse-engineering agent on the pi coding agent — analyzes apk/.o/.hex/exe with a self-provisioning toolchain (radare2 · Ghidra · angr · jadx). Static analysis. Claude Code & Codex compatible.

TornadoRevC2★ 34

A lightweight, modular post-exploitation framework for Linux and Windows. Manages interactive reverse/bind shell sessions over TCP/TLS/mTLS and provides plugins for enumeration, execution, pivoting, and operational tasks without deploying a separate agent or dropping enumeration binaries on target.

TornadoC2★ 34

Pull-based C2 framework with ten native transports, concurrent multi-profile handlers, and end-to-end ECDSA-signed tasking. Single Go binary agent with every property baked in at build time. Multi-operator team server with RBAC, audit chain, and a Qt/CLI console. Linux + Windows, amd64 + arm64.

trilane★ 33

TriLane: autonomous AI pentester and multi-agent security auditor for authorized gray-box testing of web apps and APIs. Built with Rust and Tauri.

XSS-AGENT★ 32

Autonomous AI C2

Hawx-Recon-Agent★ 32

Hawx Recon Agent is an intelligent, autonomous reconnaissance system powered by a Large Language Model (LLM). Designed for offensive security workflows, it automates initial triage and guided follow-up based on live service data.

AgenticRed★ 32

An automated pipeline that leverages LLM's meta-learning capability to iteratively design and refine red-teaming systems without human intervention.

CyberPenda★ 32

Local-first pentest agent: Go daemon + React dashboard + Codex/Claude Code/Pi runtimes, scoped projects, Blackboard, sandbox runners

pentest-agent-system★ 31

The Pentest Agent System is an autonomous penetration testing framework built on the MITRE ATT&CK framework.

Pentesting-AI★ 31

Multi agent penetration testing AI

MASAPT★ 31

Multi-Agent System for Automated Penetration Testing (MASAPT)

Hex-★ 31

Hex is a cybersecurity AI assistant for ethical hackers and red teamers. Powered by DeepSeek LLM, it simulates real-world attacks, crafts payloads, and analyzes recon data — all in a controlled, legal context. Built for speed, precision, and stealth.

drakben★ 31

Autonomous AI pentesting agent that understands natural language. Self-healing, self-evolving, multi-LLM powered.

exploithunter-agent-suite★ 31

ExploitHunter v2026 is a browser-based web security research suite for offensive testing, penetration testing, and lab assessment, combining chat-oriented workflows, LLM-supported agents, and built-in reporting.

JGD★ 31

Autonomous agent that mines Java deserialization gadget chains from any JAR directory. Static bytecode analysis + dynamic JVM probes + dual-model adversarial auditing → weaponized PoC. Solving the last mile.

llm-redteam-pipeline★ 31

An automated, multi-turn LLM red-teaming pipeline that pits an Attacker model against a Target model to evaluate safety boundaries, featuring LLM-as-a-Judge scoring and an interactive HTML vulnerability dashboard.

RedAGPT★ 30

An experimental open-source AutoGPT and Langchain based tool kit for network security

voucher_swap★ 30

Obtains the kernel task port and establishes a kernel function calling primitive on the iPhone XS, iPhone XR, and iPhone 8 running iOS 12.1.2

ghost★ 30

👻 ghost gpt– A powerful Termux-based penetration testing suite for ethical hackers. Includes free and premium tools to enhance your security testing workflows.

Strix-Kali-AI-PoweredPentestFramework★ 30

🦅Strix 落地与调教完全指北 —— 这是一个专注于下一代自主 AI 渗透测试代理Strix 的实战部署与性能榨干指南。🎮以及如何通过 --instruction 指令化身“赛博御兽师”操纵 AI 黑客注意力的进阶玩法的核心沉淀。⚡内含国内网络环境下 Docker 0/33 沙箱卡死究极加速方案,💰全网超全的多模型(DeepSeek-V4、Claude、火山引擎、OpenRouter)高性价比接入配置。A comprehensive hands-on guide to maximizing Strix, the autonomous AI agent that thinks and acts like a real hacker.

Kryon★ 30

Autonomous offensive-security agent: recon, authorized pentest, vuln hunting, DFIR & multi-framework compliance from a single prompt. Local-first (llama.cpp), skill-based, deterministic-first.

More in Security & AI safety