OHI // OPEN-SOURCE OFFENSIVE AI

Offsec Harness Index

An attempt to track the open-source offensive AI agent harnesses we can verify on GitHub, and to show which ones attackers are known to have run in real operations. Each attack row links the vendor, IR-firm, CERT or government report it rests on. We will have missed some, and corrections are welcome.

CONFIRMEDShown in a real operation OBSERVEDAimed at real targets NO RECORDNo attack in primary reporting
TRACKED154Open-source offensive AI harnesses tracked
IN THE WILD14Seen in real attacks: 9 confirmed, 5 observed
Confirmed 9Observed 5
ACTORS26Distinct attacker groups or operators
SOURCES22Publishers behind 32 primary reports, listed in section 03
BREACHES7Tied to a confirmed breach: ARTEX, Cairn, CyberStrikeAI, PentestGPT, PentAGI, OpenClaw, Hermes Agent

01 // EVIDENCESeen in real attacks

Sorted by strength of evidence
Every row links its primary source

Attacker What it did
01 ARTEXmulti-agent pentest platform; repo closed Oct 2026, forks live onReconExploitationPost-exploitation CONFIRMED VAPOR PANDAChina-nexus espionageStorm-0940China-state aligned
+5 more actorsStorm-3358Cambodia-based, targets financial firmsCRAFTY FALCONtargets Indian financial firmsUnnamed actorChinese-speaking, financially motivatedOnyx SleetNorth Korean state actor (Microsoft)Pearl SleetNorth Korean state actor (Microsoft)
One controller hit 200+ organizations in 11 days, with access confirmed at nine. Storm-0940 used it to try sign-ins at 2,250+ Microsoft 365 tenants. Zscaler tied a theft of personal data on about 67,000 people from South Korean financial firms to it. Check Point read operator consoles: one operator stole 300+ iOS code-signing certificates. GMO found it on the attack infrastructure behind breaches of Japanese companies. Microsoft names two North Korean actors using it. 200+organizations targeted Yes
2026-10 · 11
02 Cairnmulti-agent pentest frameworkReconExploitationPost-exploitation CONFIRMED CL-STA-1238likely China-nexus espionageCRAFTY FALCONtargets Indian financial firms
+3 more actorsRetail card-theft operatorunnamed, financially motivated, prompts in ChineseAIA0001NorForge cluster: Chinese-language, e-commerce accessAIA0002NorForge cluster: English-language, card and crypto theft
Given a domain and a one-line goal, ran 25 operations against at least 19 organizations, with access or data taken at 10. Gambit counted 105 attack projects launched between 10 and 15 September and at least 27 companies compromised; NorForge found 20+ exposed panels and confirmed a live uploader web shell at a hardware-parts vendor. Google separately saw it used against a US aerospace target. 27+companies compromised in one campaign Yes
2026-09 · 11
03 CyberStrikeAILLM orchestrator with a built-in C2ReconScanningVuln discoveryExploitationPost-exploitation CONFIRMED Unattributed operatorsKaspersky could not link them to a known groupCRAFTY FALCONtargets Indian financial firms
+1 more actorUnattributed operator (Japan)TrendAI: Simplified-Chinese console, targets in Japan
Implant deployed on education-sector networks in Latin America and Africa. The operator set the plan; the model on the command server chose and issued the technical steps: credential dumping, lateral movement, DCSync and Golden Ticket attacks, then database export. TrendAI separately confirmed it running 44 agents against at least 19 organizations, 16 or more in Japan, in October; deep compromise reached only a small share. Team Cymru found its banner on an IP Amazon had published from an AI-augmented actor's infrastructure, with NetFlow to the FortiGate devices that actor targeted. 3–4 daysto domain admin Yes
2026-03 · 11
04 PentestGPTautonomous pentest agentReconScanningVuln discoveryExploitationResearch CONFIRMED UAT-10147Chinese-speaking, financially motivatedUnnamed operatorSpanish- and English-speaking Installed on the attacker’s command server, scanned web servers and broke into at least one site. An operator running a rebuilt fork compromised at least 15 Argentine healthcare organizations. 170,000URLs targeted Yes
2026-08 · 10
05 PentAGImulti-agent pentest systemScanningExploitationPost-exploitation CONFIRMED GTG-50020Russian-speaking criminalsGTG-50029French-speaking hacktivistCRAFTY FALCONtargets Indian financial firms Named by Anthropic as a framework both groups used. One stole production AI API keys; the other dumped European political databases. Google saw most PentAGI activity stop at scanning, but some reach intrusion and persistence at multiple organizations. 12–26 GBdatabases taken Yes
2026-09 · 11
06 OpenClawgeneral-purpose agentExploitationPost-exploitation CONFIRMED Unnamed operatorSpanish-speaking“Recon” operatorfinancially motivated An agent on OpenClaw exploited a Telegram Mini App, dumped its database of 1,300+ users and verified the app's bot token. Google also found an .openclaw/ directory on a credential-harvesting server holding 23,800+ secrets. 1,300+user records dumped Yes
2026-08 · 09
07 Hermes Agentopen-source autonomous agent with memory and self-written skillsReconScanningExploitationPost-exploitation CONFIRMED knaithe / KnYuanChinese-speakingRetail card-theft operatorunnamed, financially motivated, prompts in Chinese
+2 more actorsSilent LynxCentral Asia espionage cluster, medium-confidence Kazakhstan linkBOSSAI operatorlone operator; Acronis found no clear affiliation
Four operations. Unit 42: exploit attempts against Langflow and n8n failed and a Citrix exploit run by hand got in. Gambit: the operator's console in a retail card-theft campaign. Acronis: the main attack agent in a lone operator's 12-agent mesh, extracting data at a Thai university. CloudSEK: handled recon, collection and log cleanup for an espionage cluster inside Tajikistan’s railway network. 4operations reported Yes
2026-07 · 11
08 oh-my-claudecodeadd-on layer for Claude CodeScanning CONFIRMED Zerofot Ran the IT and DevOps for a credential harvester its operator built with Codex and Claude Code: brought up, restarted and debugged the live scanner and its proxies. 1,000shosts harvested Tooling
2026-08-31
09 Strixmulti-agent pentest frameworkReconScanningVuln discoveryExploitationExploit devPost-exploitation CONFIRMED Suspected China-linked actorplus an attacker on hijacked AI serversRetail card-theft operatorunnamed, financially motivated, prompts in Chinese In GTIG's suspected China-linked campaign, validated vulnerabilities. A honeypot caught it aimed at a live French shop with no permission prompts. Gambit saw it run 146 times against 138 hosts as the vulnerability-search stage of a retail card-theft operation, and some of its reports opened the next stage of exploitation, handed to Cairn. 2,000+unattended steps Discovery only
2026-05 · 09
10 HexStrike AIMCP server wrapping 150+ offensive toolsReconScanningExploit devResearch OBSERVED Suspected China-linked actorthe same actor as the Strix caseBOSSAI operatorlone operator; Acronis found no clear affiliation Moved between recon tools on its own against a Japanese tech firm and an East Asian security company. Google listed it among frameworks abused in incident response in October, then described what it had seen of HexStrike as experimentation in November. Acronis found it installed as an MCP tool server on a lone operator's attack node and wired into every agent that needed it; the report ties no breach to it. 2named targets Not stated
2026-05 · 11
11 DeepAuditmulti-agent source-code auditorReconScanningVuln discoveryExploit dev OBSERVED UAT-10147Chinese-speaking, financially motivated Installed on the attacker's management server, and named among the frameworks the actor used to automate intrusions. Talos did not see it used against victims. 1attacker server Not stated
2026-08-20
12 AntiFoxARTEX derivative; Chinese-language agentic pentest platformExploitationPost-exploitation OBSERVED Unattributed AntiFox operatorsArctic Wolf names no actor; AntiFox hosted on Alibaba Cloud Infrastructure hosting AntiFox on Alibaba Cloud was “associated with” confirmed intrusions: a web exploit, an IIS web shell, AWS credential checks and a successful read of a production secret. Arctic Wolf ties the activity to the infrastructure, not to the tool itself. 1+production secret read Associated
2026-11-10
13 CyberStrikeopen-source AI pentest agent; not CyberStrikeAIReconVuln discoveryExploitation OBSERVED BOSSAI operatorlone operator; Acronis found no clear affiliation One agent in a lone operator's 12-agent attack mesh. Acronis says it appears to have run vulnerability assessments and exploitation attempts against live targets, but its outputs could not be verified. 1 of 12agents in the operator's mesh Not stated
2026-11-04
14 HalberdStrikepentest harness based on PentestGPTReconVuln discovery OBSERVED BOSSAI operatorlone operator; Acronis found no clear affiliation Mapped the attack surface across at least 20 projects, from a Taiwanese emergency-services portal to 50+ commercial sites, and passed its reports to other agents for exploitation. Over half its high and critical findings were false positives or misclassifications. 20+projects mapped Not stated
2026-11-04

02 // FIELDNo attack on record

140 more with no attack on record
GitHub facts as of 2026-11-17

What it is Repo
01 ShannonKeygraphHQ/shannonReconVuln discoveryExploitationExploit dev Pentest agent AI pentester for web apps and APIs. Reads the source, then attacks the running app to prove each finding. 52,013 AGPL-3.0 2026-11-16 GitHub ↗
02 reverse-skillzhaoxuya520/reverse-skillExploit devResearch Skill pack Skill router that sends AI coding agents to reverse-engineering, pentest and CTF toolchains. 44,828 MIT 2026-11-14 GitHub ↗
03 Agent Zeroagent0ai/agent-zeroNo offensive use stated in its README General agent General agent framework that writes and runs its own code and tools. Google saw threat actors experimenting with it (GTIG ↗). 19,900 MIT 2026-11-17 GitHub ↗
04 Claude-RedSnailSploit/Claude-RedReconVuln discoveryExploitationExploit devPost-exploitation Skill pack Library of offensive security skills for the Claude skills system. 11,173 MIT 2026-11-17 GitHub ↗
05 Claude-BugHunterelementalsouls/Claude-BugHunterReconVuln discoveryExploitationPost-exploitation Skill pack Claude Code skill bundle for bug hunting and external red-team work. 8,917 MIT 2026-11-17 GitHub ↗
06 T3MP3STelder-plinius/T3MP3STReconScanningVuln discoveryExploitationPost-exploitation Red-team platform Autonomous red-teaming platform: a multi-agent offensive meta-harness. 7,070 AGPL-3.0 2026-09-08 GitHub ↗
07 DecepticonBitterSecurity/DecepticonReconVuln discoveryExploitationPost-exploitation Red-team platform Autonomous hacking agent for red teams. 6,634 Apache-2.0 2026-11-17 GitHub ↗
08 Agentic-Bug-Hunterawarexone/Agentic-Bug-HunterReconScanningVuln discoveryExploitationExploit dev Bug-bounty agent AI bug-bounty hunting toolkit. 5,968 MIT 2026-11-12 GitHub ↗
09 METATRONsooryathejas/METATRONReconScanningVuln discovery Pentest agent Local-LLM pentest assistant that runs recon tools and has the model analyze the results. 4,281 MIT 2026-04-11 GitHub ↗
10 RAPTORgadievron/raptorVuln discoveryExploit dev Code auditor Claude Code set up as an offensive and defensive security research agent. 4,032 MIT 2026-11-17 GitHub ↗
11 VulnClawNetw0rkNoob/VulnClawReconVuln discoveryExploitationExploit dev Pentest agent Chinese-language pentest CLI agent that takes a target from recon through exploitation to a report. 3,937 MIT 2026-10-30 GitHub ↗
12 ctf-skillsljagiello/ctf-skillsResearch Skill pack Agent skills for solving CTF challenges across web, pwn, crypto, reversing and forensics. 3,730 MIT 2026-11-15 GitHub ↗
13 RedAmonsamugit83/redamonReconScanningExploitationPost-exploitation Pentest agent Self-hosted pentest framework that maps the attack surface into a graph and works it autonomously. 3,428 MIT 2026-11-14 GitHub ↗
14 PentestAgentGH05TCREW/pentestagentReconScanning Pentest agent Agent framework for black-box testing: bug bounty, red team and pentest. 3,307 MIT 2026-11-09 GitHub ↗
15 Pentest-Swarm-AIArmur-Ai/Pentest-Swarm-AIReconScanningVuln discoveryExploitationExploit dev Pentest agent Swarm of agents that runs recon, classification and exploitation. 3,248 AGPL-3.0 2026-11-13 GitHub ↗
16 Claude-OSINTelementalsouls/Claude-OSINTRecon Skill pack Eight Claude skills that prime the model with offensive recon methodology: asset discovery and exposure checks. 3,067 MIT 2026-10-20 GitHub ↗
17 Vulnhuntrprotectai/vulnhuntrVuln discoveryExploit dev Code auditor Zero-shot vulnerability discovery with LLMs: traces code paths to find remotely exploitable bugs. 2,824 AGPL-3.0 2025-02-06 GitHub ↗
18 hack-skillsyaklang/hack-skillsPost-exploitation Skill pack Knowledge base of 102 hacking skills that AI agents load, covering web, AD, pwn and mobile. 2,667 MIT 2026-09-13 GitHub ↗
19 pentest-ai-agents0xSteph/pentest-ai-agentsReconScanningVuln discoveryExploitationExploit devPost-exploitation Skill pack Offensive-security subagents for Claude Code. 2,455 MIT 2026-10-06 GitHub ↗
20 Open KrittKritt-ai/open-krittVuln discovery Code auditor Self-hosted platform that orchestrates AI agents to find and validate vulnerabilities in code. 2,430 AGPL-3.0 2026-11-16 GitHub ↗
21 Deep Eyezakirkun/deep-eyeVuln discovery Pentest agent Pentest tool that has LLMs generate payloads and scan targets for 50+ vulnerability types. 2,375 MIT 2026-10-30 GitHub ↗
22 sapperN1nEmAn/sapper-v2Research Pentest agent Multi-agent pentest orchestration harness that won a Tencent Cloud attack-defense challenge qualifier. 2,305 MIT 2026-11-11 GitHub ↗
23 emp3r0rjm33-m0/emp3r0rPost-exploitation C2 with MCP Linux-first post-exploitation C2 whose MCP server lets an AI harness operate it in natural language. 2,112 GPL-3.0 2026-11-17 GitHub ↗
24 pentest-ai0xSteph/pentest-aiVuln discoveryExploitation Pentest agent AI pentester that re-runs each exploit to prove the finding. 1,841 MIT 2026-11-16 GitHub ↗
25 Burp AI Agentsix2dez/burp-ai-agentScanningVuln discovery MCP server Burp Suite extension that adds AI scanners and an MCP bridge for AI agents. 1,594 MIT 2026-10-21 GitHub ↗
26 Pentest Copilotbugbasesecurity/pentest-copilotReconScanningExploitation Assistant Browser-based AI assistant for ethical hacking. 1,563 MIT 2026-08-14 GitHub ↗
27 NeuroSploitJoasASantos/NeuroSploitReconScanningVuln discoveryExploitationExploit devPost-exploitation Pentest agent AI-powered penetration testing framework. 1,538 MIT 2026-11-16 GitHub ↗
28 LuaN1aoAgentSanMuzZzZz/LuaN1aoAgentVuln discoveryExploitation Pentest agent Fully autonomous pentest agent built on graph-based reasoning. 1,486 AGPL-3.0 2026-09-15 GitHub ↗
29 PentesterFlowPentesterFlow/agentReconVuln discovery Pentest agent Terminal pentest agent with skills, a Burp bridge and coverage tracking that keeps the analyst in control. 1,457 Apache-2.0 2026-08-31 GitHub ↗
30 recon-skillsuphiago/recon-skillsRecon Skill pack Skill pack for agents covering external recon, web and API testing, validation and attack paths. 1,385 MIT 2026-11-07 GitHub ↗
31 Xalgorixxalgorix/xalgorixReconScanningVuln discoveryExploitation Pentest agent Pentest agent that exploit-verifies every finding across a 22-phase run. 1,366 Apache-2.0 2026-11-17 GitHub ↗
32 hackingBuddyGPTipa-lab/hackingBuddyGPTScanningExploitationPost-exploitationResearch Research framework Research framework for building LLM hacking agents in 50 lines or less. 1,333 MIT 2026-10-24 GitHub ↗
33 Dark-MoonASCIT31/Dark-MoonReconScanningVuln discoveryExploitation Pentest agent Autonomous pentesting with 50 specialist agents across web, API, cloud and Active Directory. 1,259 GPL-3.0 2026-10-27 GitHub ↗
34 AutoCVElarlarua/AutoCVEVuln discovery Code auditor Chinese-language agent platform that audits source code for vulnerabilities and drafts CVE reports. 1,229 AGPL-3.0 2026-10-01 GitHub ↗
35 VulnHunter (Capital One)capitalone/VulnHunterVuln discoveryExploit dev Code auditor Agentic source-code hunter that reasons like an adversary and drops findings it cannot prove exploitable. 1,180 Apache-2.0 2026-10-07 GitHub ↗
36 Nebulaberylliumsec/nebulaScanning Assistant Cybersecurity harness for AI-assisted engagements. 1,163 BSD-2-Clause 2026-11-17 GitHub ↗
37 Vigoliumvigolium/vigoliumScanningExploit dev Pentest agent Vulnerability scanner with an agentic mode whose swarm picks modules and writes custom attack extensions. 1,151 MIT 2026-11-16 GitHub ↗
38 codex-redteam-modechAng-L19/codex-redteam-modeThe README describes a red-team runtime and a CTF instruction file but does not name a specific offensive use. General agent Opt-in red-team operation runtime for Codex that injects a jailbreak prompt into the system layer. 1,137 MIT 2026-10-09 GitHub ↗
39 PentestCodes0ld13rr/pentestcodeReconScanningVuln discoveryExploitationExploit devPost-exploitation Pentest agent Multi-agent pentest system that keeps engagement state between runs. 1,108 MIT 2026-11-05 GitHub ↗
40 AIReconpikpikcu/aireconReconScanningVuln discoveryExploitationExploit dev Pentest agent Autonomous cybersecurity agent built on a self-hosted LLM through Ollama. 1,094 MIT 2026-11-12 GitHub ↗
41 ClearwingLazarus-ai-OSS/clearwingReconScanningExploitationExploit dev Pentest agent Autonomous scanner and vulnerability hunter that runs Kali tools and builds exploits with human approval. 1,090 MIT 2026-11-05 GitHub ↗
42 CybermesZyrexnn/CybermesReconScanningVuln discoveryExploitationExploit dev Bug-bounty agent Offensive security and bug-bounty agent framework built on Hermes Agent. 1,038 Apache-2.0 2026-11-16 GitHub ↗
43 Z3r0yv1ing/Z3r0Vuln discovery Red-team platform AI red-team workbench with specialist agents, sandboxed tooling and replayable timelines. 1,025 MIT 2026-09-13 archived GitHub ↗
44 Caincdxiaodong/cain-agentReconScanningVuln discovery Pentest agent AI pentest engineer for real authorized assessments, with 46 tools and a cloud penetration module. 953 Apache-2.0 2026-10-14 GitHub ↗
45 ctf-agentverialabs/ctf-agentResearch CTF agent Multi-model swarm that solves CTF challenges autonomously; won first place at BSidesSF 2026. 939 MIT 2026-03-29 GitHub ↗
46 dsh-redteam-modelSeaOf0/dsh-redteam-modelVuln discoveryExploitation Skill pack Ten red-team working modes and runtime plugins for the DeepSeek harness; README in Chinese. 897 MIT 2026-10-27 GitHub ↗
47 IoTHackBotBrownFineSecurity/iothackbotReconVuln discovery Skill pack IoT security testing toolkit with Claude Code skills for vulnerability discovery. 890 MIT 2026-06-01 GitHub ↗
48 auditevilsocket/auditVuln discoveryExploit dev Code auditor Eight-stage Claude agent pipeline that hunts vulnerabilities and compiles proof-of-concept code. 889 MIT 2026-06-10 GitHub ↗
49 Ars0n Framework v2R-s0n/ars0n-framework-v2ReconScanning Bug-bounty agent Bug bounty recon framework whose MCP server exposes data and scans to AI assistants. 883 GPL-3.0 2026-11-16 GitHub ↗
50 MCP-Kali-ServerWh0am123/MCP-Kali-ServerReconExploitation MCP server MCP setup that gives an AI agent a Kali Linux machine and its tools. 879 MIT 2026-03-17 GitHub ↗
51 numasecFrancescoStabile/numasecReconScanning Pentest agent AI agent for cyber security work. 842 AGPL-3.0 2026-05-08 GitHub ↗
52 FofaMapasaotomo/FofaMapReconScanning Recon agent Chinese tool that ties FOFA asset discovery, AI reflection and Nuclei validation into one workflow. 769 Apache-2.0 2026-11-02 GitHub ↗
53 MetasploitMCPGH05TCREW/MetasploitMCPScanningExploitationExploit devPost-exploitation MCP server MCP server that lets an agent drive Metasploit. 765 Apache-2.0 2026-02-05 GitHub ↗
54 StrikeAgentYean-Sec/StrikeAgent_AtkBrain-FlashVuln discovery Pentest agent Chinese AI pentest platform with red-team re-rating and re-verification of findings, tested in CTF settings. 745 AGPL-3.0 2026-11-13 GitHub ↗
55 MutekiFishCodeTech/mutekiExploitation Pentest agent Offensive security harness for automated penetration testing. 687 AGPL-3.0 2026-11-14 GitHub ↗
56 src-hunter-skillMyuriKanao/src-hunter-skillVuln discovery Skill pack Chinese Claude Code skill for SRC and bug bounty hunting with attack playbooks and payloads. 656 MIT 2026-05-24 archived GitHub ↗
57 MCP for Securitycyproxio/mcp-for-securityReconScanning MCP server MCP servers for SQLMap, FFUF, Nmap, Masscan and other security tools, for AI pentest workflows. 632 MIT 2026-03-30 archived GitHub ↗
58 communitytoolstransilienceai/communitytoolsReconScanningVuln discoveryExploitationExploit devPost-exploitation Skill pack Claude Code skills, agents and commands for pentesting and bug bounty. 624 MIT 2026-07-29 GitHub ↗
59 CHYing-agentyhy0/CHYing-agentResearch CTF agent Agent from Tencent Cloud AI pentest hackathons, ninth in the first year; README in Chinese. 596 MIT 2026-04-25 GitHub ↗
60 Cyber-AutoAgentwestonbrown/Cyber-AutoAgentReconScanningExploitationExploit dev Pentest agent AI agent for autonomous cyber operations. 552 MIT 2025-11-29 archived GitHub ↗
61 MitsuARCANGEL0/MitsuReconVuln discoveryPost-exploitation Pentest agent Pentest agent client that runs an authorized engagement from recon through post-exploitation with 165+ MCP tools. 541 MIT 2026-11-17 GitHub ↗
62 Flounderadshao/flounderVuln discoveryExploit dev Bug-bounty agent Turns coding agents into a security auditor for bug bounty work, with local exploit construction. 527 AGPL-3.0 2026-11-15 GitHub ↗
63 Blitz Strikeshinthink/blitzstrikeReconVuln discoveryExploitation MCP server MCP security toolbelt that maps attack surface, traces source-to-sink paths and verifies findings live. 518 MIT 2026-11-04 GitHub ↗
64 AIDAVasco0x4/AIDAReconScanningVuln discoveryExploitationExploit dev Pentest agent Turns any LLM into an autonomous pentester inside a scope you set. 515 AGPL-3.0 2026-10-21 GitHub ↗
65 Mantishackdeonmenezes/mantishackVuln discovery Code auditor AI vulnerability pipeline with recon, attacker-simulation validation and gated exploitation over a codebase. 507 Apache-2.0 2026-10-03 GitHub ↗
66 Zen-Ai-PentestSHAdd0WTAka/Zen-Ai-PentestReconScanningExploitationPost-exploitation Pentest agent Multi-agent pentest framework with automated vulnerability scanning. 506 MIT 2026-11-03 GitHub ↗
67 AdStrikecapture0x/AdStrikeExploitationPost-exploitation Pentest agent Active Directory attack framework with an MCP server so the user's own LLM host drives 53 tools. 489 MIT 2026-10-17 GitHub ↗
68 Roguefaizann24/rogueReconVuln discovery Pentest agent LLM web vulnerability scanner agent that generates test payloads and verifies findings. 483 GPL-3.0 2025-06-18 GitHub ↗
69 Pentest HarnessS1N6H/pentest-harnessThe README is generic: it describes an offensive-security operating standard and names pentests, bug bounty and CTFs without a specific technique. Pentest agent Self-hosted AI agent harness for authorized pentests, bug bounty, security labs and CTFs. 474 MIT 2026-10-07 GitHub ↗
70 BoxPwnr0ca/BoxPwnrResearch CTF agent Framework that benchmarks LLM agents solving HackTheBox, PortSwigger, CTF and other security labs. 471 AGPL-3.0 2026-07-24 GitHub ↗
71 Bug-Bounty-Agentsmatty69v/Bug-Bounty-AgentsReconVuln discoveryExploit devPost-exploitation Skill pack Drop-in agent personas for bug bounty recon, web hunting, PoC validation and reporting. 470 MIT 2026-04-30 GitHub ↗
72 OpenHackopenhackai/OpenHackVuln discovery Code auditor Open-model agent scanner that hunts, validates and exploit-tests vulnerabilities in a codebase. 445 MIT 2026-10-17 GitHub ↗
73 cyber-harnesschainreactors/cyber-harnessReconScanningVuln discovery Pentest agent Single-binary agent for pentest, red team and bug bounty. 422 AGPL-3.0 2026-11-17 GitHub ↗
74 offensive-claudehypnguyen1209/offensive-claudeReconVuln discoveryExploit devPost-exploitation Skill pack Spec-driven offensive framework for Claude Code with 32 kill-chain skills. 421 MIT 2026-10-22 GitHub ↗
75 RedCellmartian56/redcellScanningVuln discoveryExploitationPost-exploitation Pentest agent Agents run a penetration test end to end inside a Kali container. 414 MIT 2026-11-12 GitHub ↗
76 BugTraceAIBugTraceAI/BugTraceAIReconScanningVuln discoveryExploit dev Pentest agent Self-hosted web pentesting suite with agentic scanning. 389 Apache-2.0 2026-10-19 GitHub ↗
77 Vibe Pentestok-helloworld/vibe-pentestVuln discovery Pentest agent Chinese multi-agent black-box pentest tool that mimics a human expert, including business-logic flaws. 387 AGPL-3.0 2026-09-21 GitHub ↗
78 Halberdvectra-ai-research/HalberdPost-exploitation Red-team platform Multi-cloud attack emulation tool with a Claude-powered agent that helps discover and execute attack paths. 375 GPL-3.0 2026-04-08 GitHub ↗
79 pentest-skillscrazyMarky/pentest-skillsRecon Skill pack Chinese pentest skill set for Claude Code and Gemini CLI that turns natural language into tool runs. 368 Apache-2.0 2026-06-04 GitHub ↗
80 h1-brainPatrikFehrenbach/h1-brainRecon MCP server MCP server linking an AI assistant to HackerOne bug bounty history, scopes and reports. 360 MIT 2026-10-13 GitHub ↗
81 RedAIkpolley/redaiVuln discoveryExploit dev Pentest agent Terminal workbench that finds vulnerabilities with AI, then validates them live with PoC scripts. 356 MIT 2026-05-05 GitHub ↗
82 TARSosgil-defense/TARSScanning Pentest agent Agents that automate pentesting. 351 MIT 2025-01-17 archived GitHub ↗
83 Pojunchu0119/PojunReconVuln discoveryExploitationPost-exploitation Red-team platform Chinese multi-agent platform that takes a one-line goal through recon, exploitation and post-exploitation to a report. 349 AGPL-3.0 2026-11-17 GitHub ↗
84 DorkAgentyee-yore/DorkAgentRecon Recon agent LLM agent that automates Google dorking for bug hunting and pentesting. 341 MIT 2026-08-24 GitHub ↗
85 LeoAIcha0upup/LeoAIPost-exploitation C2 with MCP Chinese post-exploitation management platform with platform-level and per-node AI copilots. 333 GPL-3.0 2026-11-17 GitHub ↗
86 deadend-clistraylabs-ai/deadend-cliReconVuln discoveryExploitationExploit dev Pentest agent Self-hosted agentic pentest tooling for black-box testing, on any LiteLLM model. 331 AGPL-3.0 2026-08-07 GitHub ↗
87 LyrieOTT-Cybersecurity-LLC/lyrie-aiReconScanning Pentest agent Pentest CLI that fingerprints and scans targets with AI-assisted exploit validation; also red-teams LLM endpoints. 329 MIT 2026-11-17 GitHub ↗
88 Apexpensarai/apexVuln discovery Pentest agent Terminal pentest agent that runs blackbox and whitebox tests with agent swarms in a Kali container. 327 Apache-2.0 2026-11-17 GitHub ↗
89 HackSynthaielte-research/HackSynthResearch Research framework LLM agent and evaluation framework for autonomous penetration testing. 324 AGPL-3.0 2025-06-24 GitHub ↗
90 PlamenPlamenTSV/plamenVuln discoveryExploit dev Code auditor Claude Code and Codex agent swarm that audits smart contracts and node clients with PoC exploits. 316 MIT 2026-09-26 GitHub ↗
91 red-runblacklanternsecurity/red-runReconPost-exploitation Skill pack Claude Code skills, MCP servers and agent teams for pentest phases from recon to privilege escalation. 309 AGPL-3.0 2026-11-12 archived GitHub ↗
92 BurpMCP-UltraCy-S3c/BurpMCP-UltraScanningVuln discovery MCP server MCP server giving AI agents 150 Burp Suite Professional tools for scanning, fuzzing and traffic analysis. 309 MIT 2026-09-28 GitHub ↗
93 LazyOwngrisuno/LazyOwnReconScanningExploitationExploit devPost-exploitation C2 with MCP Multi-operator C2 that exposes 153 MCP tools to AI agents. 302 AGPL-3.0 2026-11-17 GitHub ↗
94 Scan-Spring-GOsspsec/Scan-Spring-GOExploitation MCP server Go scanner and exploiter for Spring components that exposes all its modules to AI clients over MCP. 302 MIT 2026-08-29 GitHub ↗
95 WolfShell0x7556/wolfshellPost-exploitation C2 with MCP ASPX WebShell and C2 manager with an MCP server and AI-assisted penetration. 286 MIT 2026-08-16 GitHub ↗
96 vulnerator-aiirraz/vulnerator-aiScanning Pentest agent OODA-loop pentest agent on Hermes, with Kali and Playwright tools over MCP. 285 MIT 2026-11-17 GitHub ↗
97 SecGPTZacharyZcR/SecGPTScanningVuln discoveryResearch Pentest agent AutoGPT-style LLM tool for pentesting, vulnerability scanning and code audit, with CTF aims. 285 Apache-2.0 2023-11-15 GitHub ↗
98 raptor-loop-huntdinosn/raptor-loop-huntVuln discovery Code auditor Claude Code skill that loops a multi-altitude vulnerability hunt over a codebase. 278 MIT 2026-09-01 GitHub ↗
99 BlacksmithAIyohannesgk/blacksmithReconVuln discoveryExploitationPost-exploitation Pentest agent Multi-agent pentest framework with separate recon, vulnerability, exploit and post-exploit agents. 269 AGPL-3.0 2026-10-23 GitHub ↗
100 AutoPentest AIbhavsec/autopentest-aiReconVuln discoveryExploitation MCP server MCP server where Scout, Analyzer, Exploiter and Reporter agents run web pentests against OWASP test cases. 251 Apache-2.0 2026-02-22 GitHub ↗
101 Haiou Workshophaiou-666/Haiou-workshopResearch Assistant Chinese AI workbench for reverse engineering, authorized pentests and CTF using Codex and Claude Code. 237 MIT 2026-10-25 GitHub ↗
102 VulnBotKHenryAegis/VulnBotResearch Research framework Multi-agent framework for autonomous penetration testing. 236 MIT 2026-11-08 GitHub ↗
103 Watchtowerfzn0x/watchtowerReconScanning Pentest agent LangGraph pentest CLI where LLM agents orchestrate 28 security tools. 223 MIT 2026-11-15 GitHub ↗
104 Claude-ADADScanPro/Claude-ADReconPost-exploitation Skill pack Active Directory pentest playbook for Claude Code as skills, agents and slash commands. 218 MIT 2026-08-24 GitHub ↗
105 web3-bug-bounty-hunting-ai-skillsawarexone/web3-bug-bounty-hunting-ai-skillsVuln discovery Skill pack Eighteen Claude Code skill files for smart contract bug bounty hunting. 192 MIT 2026-09-21 GitHub ↗
106 AIxVulnm4xxxxx/AIxVulnVuln discoveryExploit dev Code auditor Chinese LLM multi-agent system that audits code, then verifies vulnerabilities with PoCs in Docker sandboxes. 191 MIT 2026-04-25 GitHub ↗
107 kali-pentestx-glacier/kali-pentestPost-exploitation Skill pack Kali Linux pentest skill for Claude Code, OpenClaw and Hermes Agent, with 200+ CLI tools. 187 Apache-2.0 2026-06-25 GitHub ↗
108 NYU CTF agentsNYU-LLM-CTF/nyuctf_agentsResearch CTF agent Baseline and D-CIPHER LLM agents that solve CTF challenges for the NYU CTF Bench. 181 MIT 2026-07-17 GitHub ↗
109 ZypheronKKingZero/Zypheron-CLIRecon Pentest agent Local-first security CLI with recon, scanning and AI-assisted operator workflows. 174 MIT 2026-10-25 GitHub ↗
110 AI-OPSantoninoLorenzo/AI-OPSScanning Assistant Penetration testing assistant built on open-source LLMs. 172 MIT 2026-11-08 GitHub ↗
111 CkSKILLSzhaji2333/CkSKILLSReconVuln discoveryPost-exploitation Skill pack Chinese skill library that turns researcher methodology into reusable agent skills for SRC bug hunting. 159 MIT 2026-11-17 GitHub ↗
112 forefy/.contextforefy/.contextReconExploitationExploit dev Skill pack AI agent skills for security audits, pentests and bug bounty work, from recon to PoC generation. 156 MIT 2026-11-17 GitHub ↗
113 agent-smith0x0pointer/agent-smithReconExploitationPost-exploitation Pentest agent Pentest framework where skills teach the LLM a methodology and it invents the attacks itself. 155 AGPL-3.0 2026-11-14 GitHub ↗
114 cochiseandreashappe/cochiseExploitationPost-exploitationResearch Pentest agent Autonomous assumed-breach penetration testing of Active Directory networks. 153 MIT 2026-10-30 GitHub ↗
115 FuzzingBrainfuzzingbrain/afc-crs-all-you-need-is-a-fuzzing-brainVuln discoveryExploit dev Code auditor LLM system built on OSS-Fuzz that finds vulnerabilities, builds PoC inputs and patches; a DARPA AIxCC entry. 153 Apache-2.0 2026-11-17 GitHub ↗
116 Pentest MCPDMontgomery40/pentest-mcpScanning MCP server Scoped MCP server giving assessment agents nmap, Nuclei, Hashcat and other tools for authorized tests. 151 AGPL-3.0 2026-10-30 GitHub ↗
117 HeimdallQiantangCredit/heimdall-agentResearch CTF agent Self-hosted AI solver and toolkit for CTF and security labs; top 2 at two Chinese challenges. 151 MIT 2026-10-29 GitHub ↗
118 Incalmocylabcyberautonomy/IncalmoResearch Pentest agent Autonomous LLM-driven network pentest tool for multistage attacks, built for a research paper. 150 MIT 2026-11-11 GitHub ↗
119 Executionerdbx0/executionerReconScanningExploitation Pentest agent Local workspace where teams of AI agents run an assessment against your tools and scope. 147 MIT 2026-11-17 GitHub ↗
120 bloodhound_mcpmwnickerson/bloodhound_mcpRecon MCP server MCP server that lets an LLM query BloodHound attack paths across Active Directory and Entra ID. 144 GPL-3.0 2026-08-19 GitHub ↗
121 Phalanxwebxos/phalanxReconScanningExploitationPost-exploitation Pentest agent Multi-agent pentest framework with recon, exploit and post-exploit agents and 58 built-in tools. 142 MIT 2026-08-16 GitHub ↗
122 PentestAgent (nbshenxm)nbshenxm/pentest-agentReconVuln discoveryResearch Pentest agent Research LLM framework that automates intelligence gathering, vulnerability analysis and exploitation stages. 139 MIT 2025-12-20 GitHub ↗
123 claude-pentestStickman230/claude-pentestExploitationExploit devPost-exploitation Pentest agent Claude Code plugin with 15 agents and Kali MCP integration for full pentest engagements. 139 MIT 2026-06-08 GitHub ↗
124 gorihahwul/goriScanning MCP server Terminal intercepting proxy and pentest toolkit whose MCP server lets AI agents drive the same engagement. 139 Apache-2.0 2026-11-17 GitHub ↗
125 PentestAssistantInfSecLabs/PentestAssistantScanningVuln discovery Pentest agent LLM web pentest detector using function calling and multiple agents to run nmap and sqlmap. 138 Apache-2.0 2025-04-06 GitHub ↗
126 MAPTAarthurgervais/maptaNo README on GitHub to tag from Research framework Multi-agent system for autonomous web application security assessment. 136 MIT 2025-08-28 GitHub ↗
127 Moxymatank001/MoxyRecon Pentest agent Intercepting-proxy DAST tool where an AI agent reasons about attack surface, edits traffic and drives a browser. 134 Apache-2.0 2026-01-22 GitHub ↗
128 pentestkitlordx64/pentestkitReconExploitationResearch Pentest agent Multi-agent pentest framework on the Claude Agent SDK where a skeptical verifier tries to exploit each candidate. 127 MIT 2026-07-31 GitHub ↗
129 HexestraABSllk/HexestraScanning Assistant AI-native pentest workbench where Claude Code or Codex works with browser, shells and traffic capture. 119 Apache-2.0 2026-11-17 GitHub ↗
130 aimy-skillProhao42/aimy-skillExploitation Skill pack Chinese pentest skill for AI agents with a structured check interface and exploit-chain engine. 115 MIT 2026-11-09 GitHub ↗
131 Zack AI ScannerZackSecurity/Zack-AI-ScannerVuln discoveryExploitation Pentest agent Burp Suite extension that runs an LLM pipeline against one request and sends the real payloads. 112 GPL-3.0 2026-09-27 GitHub ↗
132 ZeroTracesergio11/zero_trace_hacking_toolRecon Recon agent Personal AI-powered dorking tool that uncovers sensitive information for recon. 111 MIT 2025-12-17 GitHub ↗
133 awsome-kali-MCPServersccq1/awsome_kali_MCPServersScanning MCP server Collection of MCP servers that expose Kali Linux tools to AI clients. 110 Apache-2.0 2025-11-16 GitHub ↗
134 AutoPentesterYasodGinige/AutoPentesterResearch Pentest agent Tool for automated pentesting tasks. 109 MIT 2026-09-25 GitHub ↗
135 CyberSentryprutxvi/cybersentryVuln discovery Pentest agent Autonomous AI agent for website penetration testing and vulnerability assessment. 108 MIT 2026-10-27 GitHub ↗
136 SQLMap Skynetdrcrypterdotru/sqlmap-skynetScanningExploitation Pentest agent AI-assisted SQLMap orchestration with a dashboard and an MCP tool server for agents. 108 Apache-2.0 2026-02-26 GitHub ↗
137 HEAVENnishu2402/HEAVEN-Autonomous-Penetration-TestingReconVuln discoveryPost-exploitation Pentest agent Autonomous pentest framework with an LLM plan-act loop, vuln-hypothesis agent and post-exploitation playbook. 106 MIT 2026-11-17 GitHub ↗
138 Aresdreadnode/aresExploitation Red-team platform LLM-coordinated platform whose red-team agents chain 64+ Active Directory attack tools; also has a blue-team mode. 104 MIT 2026-10-28 GitHub ↗
139 open-sesamehassan-algo/open-sesameReconVuln discovery Skill pack Claude Code pentest skills for web and network engagements, from attack-surface mapping to reports. 102 MIT 2026-10-22 GitHub ↗
140 Hacker Bobvmihalis/hacker-bobVuln discovery Bug-bounty agent Autonomous bug bounty hunter built on a local MCP runtime, with a public portfolio of reported CVEs. 101 Apache-2.0 2026-11-07 GitHub ↗

03 // SOURCESPrimary reports

32 reports from 22 publishers
Newest first

Publisher Report Cited for
2026-11-16 Google GTIGSecurity vendor GTIG AI Threat Tracker: Agentic Threat Activity Crosses a Significant Threshold ↗ ARTEX Cairn CyberStrikeAI PentAGI HexStrike AI Agent Zero
2026-11-13 Amazon Threat IntelligenceSecurity vendor Amazon Threat Intelligence observes threat actors running cybercrime campaigns with ARTEX AI agents ↗ ARTEX
2026-11-13 FBIGovernment FLASH-20261113-001: Threat Overview of ARTEX Autonomous Penetration Testing Framework ↗ ARTEX
2026-11-12 Palo Alto Networks Unit 42Security vendor Attackers Use Autonomous AI to Target Semiconductor, Defense and Civil Society Sectors ↗ Cairn
2026-11-11 Palo Alto Networks Unit 42Security vendor ARTEX and AI-Powered Red Teaming Tools: What Defenders Need to Know ↗ ARTEX
2026-11-11 CloudSEKSecurity vendor Off the Rails: Inside an Exposed Silent Lynx Server ↗ Hermes Agent
2026-11-10 Arctic Wolf LabsSecurity vendor Antifox Agentic Pentesting Platform Used to Steal AWS Credentials and Move Laterally in Compromised Environments ↗ AntiFox
2026-11-09 Datadog Security LabsSecurity vendor On the hunt for ARTEX: understanding agentic attacks in AWS environments ↗ ARTEX
2026-11-06 Microsoft Threat IntelligenceSecurity vendor ARTEX abuse expands as cybercriminals target financial sector ↗ ARTEX
2026-11-05 Check Point ResearchSecurity vendor Hacking on Autopilot: Inside ARTEX Cybercrime Campaigns ↗ ARTEX
2026-11-04 CERT-InCERT Escalation of AI-Driven Attacks on BFSI Entities by Threat Actor “CRAFTY FALCON” (CICA-2026-3651) ↗ ARTEX Cairn CyberStrikeAI PentAGI
2026-11-04 Acronis Threat Research UnitSecurity vendor BOSSAI: Inside the AI attack mesh that targeted Taiwan’s emergency services ↗ Hermes Agent HexStrike AI CyberStrike HalberdStrike
2026-11-02 Kaspersky SecurelistSecurity vendor CyberStrikeAI attacks educational institutions ↗ CyberStrikeAI
2026-10-30 Google GTIG · MandiantSecurity vendor ARTEX and Friends: Enter Agentic Offensive Frameworks ↗ ARTEX Cairn PentAGI HexStrike AI
2026-10-29 CrowdStrikeSecurity vendor VAPOR PANDA Uses AI-Driven ARTEX to Grow Global Espionage Campaign ↗ ARTEX
2026-10-29 Microsoft Threat IntelligenceSecurity vendor Tracking AI adoption in North Korean threat actor tradecraft ↗ ARTEX
2026-10-27 TrendAI (Trend Micro)Security vendor Not Only ARTEX: The Different Levels AI is Used in Attacks Targeting Japan ↗ CyberStrikeAI
2026-10-21 Microsoft Threat IntelligenceSecurity vendor AI-driven, large-scale nation-state campaign using ARTEX enables supply chain compromise ↗ ARTEX
2026-10-21 CrowdStrikeSecurity vendor CrowdStrike Uncovers New Agentic AI Campaigns ↗ PentestGPT
2026-10-19 NorForgeSecurity vendor Inside Cairn: Open Windows into massive AI-Driven Cyber Operations ↗ Cairn
2026-10-19 GMO Cybersecurity by IERAESecurity vendor 相次ぐ不正アクセスの攻撃元の多くがAIであることを確認 (in Japanese) ↗ ARTEX
2026-10-14 Zscaler ThreatLabzSecurity vendor Threat Actors Abuse AI Pentesting Framework ↗ ARTEX
2026-09-22 Gambit SecuritySecurity vendor AI Agents Are Hacking Online Retailers for $25 a Company ↗ Cairn Hermes Agent Strix
2026-09-10 AnthropicAI lab Countering misuse of AI: September 2026 ↗ PentAGI
2026-09-08 Google GTIGSecurity vendor GTIG AI Threat Tracker: From Prompting to Autonomy – The Evolution of Adversarial AI ↗ OpenClaw
2026-08-31 Gambit SecuritySecurity vendor AI Across the Intrusion Lifecycle ↗ oh-my-claudecode
2026-08-20 Cisco TalosSecurity vendor UAT-10147: Chinese-speaking adversary integrates agentic AI into post-compromise operations ↗ PentestGPT DeepAudit
2026-08-04 Cisco TalosSecurity vendor “Keep going, bro. You’ve got this!” A data-driven look at how adversaries are weaponizing AI ↗ OpenClaw
2026-07-30 Palo Alto Networks Unit 42Security vendor Chinese-Speaking Threat Actor Harnesses AI Models for Autonomous Cyberattacks ↗ Hermes Agent
2026-06-30 Zenity LabsSecurity vendor Real-World AI Infrastructure Attacks: What Honeypots Reveal ↗ Strix
2026-05-11 Google GTIGSecurity vendor Adversaries Leverage AI for Vulnerability Exploitation, Augmented Operations, and Initial Access ↗ Strix HexStrike AI
2026-03-02 Team CymruSecurity vendor Tracking CyberStrikeAI: AI-Native Offensive Tools & MSS Ties ↗ CyberStrikeAI

04 // METHODHow this is built

Evidence reviewed 2026-11-17
GitHub facts 2026-11-17

Tying a harness to an attackA harness enters table 01 only when a security vendor, IR firm, CERT, government agency or AI lab names it in its own report of a real operation, and the row links that report. We never tie a harness to an attack from overlapping TTPs, shared tooling or our own judgment. Forum posts, marketing and news stories that cite no primary report do not count. When a report hedges, the row hedges with it.

Evidence tiersConfirmed means the report shows the harness in a real operation. Observed means it was aimed at real targets or found on attacker infrastructure, with no breach stated. No attack on record means none in primary reporting, which does not prove none happened.

Use tagsOutlined tags come from each project's own README and are its claim, not a tested result. Filled tags come from attack reporting. Hover any tag to read the exact sentence behind it, click to open the README or report, or get every sentence in data.json. A tool table that lists sqlmap does not earn Exploitation; the README has to say the project attacks targets.

What the reports showIn the ARTEX and Cairn cases the frameworks ran long stretches of an intrusion on their own, mostly through old weaknesses: weak passwords, auth bypasses, known CVEs. CyberStrikeAI's agent ran textbook domain-compromise steps for a human planner, and a second operator ran 44 of its agents against organizations in Japan, with deep compromise at only a few. PentestGPT, PentAGI and the OpenClaw Telegram case also exploited live targets on their own. Hermes Agent's exploits failed in July and a human got in; in three later operations it was the attacker's console or main agent, including data extraction at a Thai university. Strix, first seen validating vulnerabilities, was the discovery stage of a retail card-theft campaign, and some of its reports went to Cairn. HexStrike, DeepAudit and HalberdStrike were seen only mapping, scanning or sitting on attacker servers. CyberStrike's outputs could not be verified, and AntiFox is tied to intrusions only through the infrastructure that hosted it.

Scope and rosterThe field table lists open-source projects under an OSI license that put an LLM in the loop of offensive security work, on public GitHub with 100 stars or more, or with a primary report that names them. The attack table needs only a primary report that names an open-source harness, at any star count, and it includes general agents threat actors were seen using. Source-available projects such as CAI are left out, and so are commercial agents like Claude Code, Codex and Gemini CLI.

The incidents ledgerOur sister site, hackingincidents.ai, records the attacks themselves: every cyberattack where AI was known or strongly suspected to be used, graded and sourced. A row here that the ledger mentions links to those incidents under what it did (9 harnesses, 12 incidents, checked 2026-11-17), and the ledger links back to the rows here.

Freshness and the dataGitHub facts and README tag sentences are re-checked on every refresh, and the refresh fails once the evidence review is older than 21 days. Each evidence pass gets an independent model review against its sources. Everything here renders from one dataset: data.json, mirrored with the full research notes in Substrate ↗. Every data source and step is laid out on the About page.