Not classifiedCyberStrikeAIHexStrike

BOSSAI 12-agent AI attack mesh targeting Taiwan emergency services

Nov 4, 2026: Acronis TRU discovered BOSSAI, a mesh of 12 AI agents (9 active, 3 inactive) operated by single actor 'bossjy' (aliases TIAN0517, lstjks) based in Hualien, Taiwan. Agents divided into…

What happened

Nov 4, 2026: Acronis TRU discovered BOSSAI, a mesh of 12 AI agents (9 active, 3 inactive) operated by single actor 'bossjy' (aliases TIAN0517, lstjks) based in Hualien, Taiwan. Agents divided into infrastructure builders (Grok, OpenClaw, Claude Code) running commercial models, and offensive agents (Hermes, HalberdStrike, CyberStrike) running abliterated local models (DeepSeek-V4, Qwen 3.5 122B). Commercial models repeatedly refused attack commands, forcing use of locally-run abliterated models. Agent mesh targeted hundreds of organizations across government emergency services, banks, universities, military systems in 11 countries. Confirmed breaches at Taiwan emergency services (patient care records, 119 dispatch data), universities (Thailand SQL injection), and financial institutions. Operator ran mesh from home IP alongside legitimate businesses (gas store, K-beauty reseller, Facebook auto-post SaaS). Attack mesh generated high volumes of data with frequent hallucinations and false positives; agents sometimes refused lucrative targets (Honduran military base) due to autonomous scope decisions. MCP servers provided GitHub security, CVE lookup, OSINT, cloud auditing, Ghidra, Radare2, Playwright automation, and Kali Linux tools (nmap, sqlmap, Metasploit, Hydra). GPU fleet: RTX 4090s serving abliterated models via llama.cpp, plus 8 VPS nodes for proxying. OPSEC failures exposed entire operation: GitHub account linked to physical storefront, SSH sessions exposed home IP, credentials hardcoded throughout.

Single operator 'bossjy' deployed mesh of 12 AI agents working together to conduct persistent espionage and intrusion campaigns across 11 countries

Read the original source

How it happened

Select a stage to explore its defenses
Select a stage · Drag to explore · + / − to zoom · Dashed edges are loopsSelect a stage · Arrows show reported connectionsReported sequence
Read the flow as text & sources

A catalog reconstruction of the supplied source account. Unknown steps remain explicit; an edge describes the reported relationship between steps.

  1. 01 · Deploy 12-agent BOSSAI mesh

    Operator bossjy deploys mesh: 3 infrastructure agents (Grok, OpenClaw, Claude Code) with commercial models; 9 offensive agents (Hermes, HalberdStrike, CyberStrike, Cursor, Trae, etc.) with abliterated models

    Source for step 1
  2. 02 · FOFA reconnaissance and target selection

    Agents scan hundreds of organizations across 11 countries; FOFA reconnaissance feeds target lists to offensive agents

    Source for step 2
  3. 03 · Autonomous vulnerability scanning

    HalberdStrike, AWVS, CyberStrike scan targets; agents generate vulnerability reports (many false positives due to hallucinations)

    Source for step 3
  4. 04 · Autonomous exploitation

    Hermes agent executes SQL injection, RCE, credential harvesting autonomously; uses MCP tools (nmap, sqlmap, Metasploit, Hydra) via Kali MCP server

    Source for step 4
  5. 05 · Data exfiltration

    Emergency medical data (Taiwan), 119 dispatch records, university student data (Thailand), 206K+ phone records (Mexico), financial data extracted

    Source for step 5

Why it matters

Reported entry point: Vulnerability exploitation (SQL injection, RCE), credential harvesting, exposed systems. Reported impact: Emergency medical data (Taiwan), 119 dispatch records, university student data (Thailand), 206K+ phone records (Mexico), financial data extracted; reported targets include Taiwan emergency services (fire department, 119 dispatch, disaster response), financial institutions (lending, mobile payment, fire-dispatch vendors), universities (Thailand: 3 confirmed SQL injection with data exfiltration, 2 partial), Mexico (206,800+ residential phone records, 124,000+ yellow pages records), Honduras military air force personnel system (characterized but not exploited), government agencies across 11 countries; applicability to your setup has not been assessed, so verify affected versions, permissions, and configuration in the original report.

Read the original report

What you can do

Protection & prevention

Monitor for autonomous AI agent behavior patterns; detect nmap, sqlmap, Metasploit execution from unexpected contexts; block MCP server abuse; restrict abliterated model access; monitor FOFA/Shodan reconnaissance patterns; detect multi-agent mesh communication; secure emergency services APIs; patch SQL injection vulnerabilities; monitor for hallucinated vulnerability reports; detect AWVS signatures

Catalog defensive guidance based on the reported behavior. Confirm scope and current mitigations in the original source.

Review the original guidance
Technical detailsTooling, impact & MITRE ATLAS

Attack context

AI's role
AI as operator
Offensive frameworks
CyberStrikeAIHexStrike
Initial access
Vulnerability exploitation (SQL injection, RCE), credential harvesting, exposed systems

Tooling as reported

AI products, models and artifacts

  • BOSSAI 12-agent mesh
  • Hermes Agent (github.com/NousResearch/hermes-agent)
  • HexStrike AI (github.com/0x4m4/hexstrike-ai)
  • PentestGPT (github.com/GreyDGL/PentestGPT)
  • HalberdStrike (PentestGPT fork)
  • CyberStrike (github.com/CyberStrikeAI/CyberStrikeAI)
  • Grok 4.5 (infrastructure)
  • Claude Code (infrastructure)
  • MiniMax M3 (infrastructure, OpenClaw agent)
  • Abliterated DeepSeek-V4 (DS4, 64K context, offensive)
  • Abliterated Qwen 3.5 122B (offensive)
  • Abliterated GLM (offensive)
  • DeepSeek-R1 70B (offensive, later replaced)

Other tooling

  • MCP servers: HexStrike AI, Kali Linux, Ghidra, Radare2, Playwright
  • nmap, sqlmap, Metasploit, Hydra
  • FOFA reconnaissance
  • Acunetix Web Vulnerability Scanner (cracked)
  • Hikvision Comprehensive Vulnerability Exploitation Tool
  • RTX 4090 GPU fleet with llama.cpp
  • SOCKS5 proxy pool
  • BrightData datacenter proxies
  • WeChat gateway
  • LINE messenger

Targets and reported impact

Targets / sectors
Taiwan emergency services (fire department, 119 dispatch, disaster response), financial institutions (lending, mobile payment, fire-dispatch vendors), universities (Thailand: 3 confirmed SQL injection with data exfiltration, 2 partial), Mexico (206,800+ residential phone records, 124,000+ yellow pages records), Honduras military air force personnel system (characterized but not exploited), government agencies across 11 countries
Reported scale / exposure
12-agent mesh; hundreds of organizations targeted across 11 countries; ~10 confirmed breaches; 206,800+ Mexican residential phone records; 124,000+ Mexican yellow pages records; Taiwan emergency medical data (ECG, ambulance video, vitals from 2018), 119 dispatch case records (7 Taipei districts); 3 Thai universities SQL injection with data extraction; 16 SQL injection points at Taiwan lending firm; ~95 Hermes offensive sessions; attack mesh ran for 12+ days under observation; 9GB evidence (90,000+ files); false positive rate >50% on high-severity findings from HalberdStrike
Target geography
Taiwan · Thailand · Mexico · Honduras · United States · United Kingdom · Australia · Sri Lanka · China · Vietnam

Reported attribution

bossjy (TIAN0517, lstjks), Hualien City, Taiwan; high confidence single operator

Structured attribution

Actor
bossjy (bossjy)
Aliases
TIAN0517 · lstjks
Country
Taiwan
Motivation
financially-motivated
Attributed by
Acronis TRU
Attribution confidence
high

Detection rules

  • Nmap/Zenmap ExecutionSIGMAView rule

    Detects nmap/zenmap execution used by BOSSAI agents for network reconnaissance

  • Linux HackTool ExecutionSIGMAView rule

    Detects execution of hacktools including nmap, sqlmap, Metasploit used by offensive agents

  • Linux Network Service ScanningSIGMAView rule

    Detects network service enumeration via auditd syscalls

  • SQL Injection in Web LogsSIGMAView rule

    Detects SQL injection attempts in web server access logs (BOSSAI targeted Thai universities with SQL injection)

  • Multi-agent AI mesh detectionGUIDANCE

    Monitor for patterns indicating multiple AI agents working in coordination: autonomous tool execution from unexpected contexts, MCP server traffic, high-volume reconnaissance, parallel targeting across multiple organizations

  • Suspicious Network Tool Launched Inside ContainerELASTICView rule

    Detects nmap, sqlmap, Metasploit tools in containers; directly applicable to BOSSAI mesh MCP-wrapped tooling

  • Web Application Suspicious Activity - sqlmap User AgentELASTICView rule

    Detects sqlmap tool used by BOSSAI against Thai universities

  • Abliterated model abuseGUIDANCE

    Detect locally-run abliterated LLM models (DeepSeek-V4, Qwen, GLM variants) serving offensive agents; monitor for llama.cpp GPU inference serving attack frameworks

Activity timeline

First seen
2026-07-01
Last updated
2026-11-05

Updates

  • 2026-11-04 — Acronis TRU

    BOSSAI 12-agent mesh disclosed; single operator bossjy targeted Taiwan emergency services and 11 countries with AI-driven attack infrastructure

Source defensive notes

Monitor for autonomous AI agent behavior patterns; detect nmap, sqlmap, Metasploit execution from unexpected contexts; block MCP server abuse; restrict abliterated model access; monitor FOFA/Shodan reconnaissance patterns; detect multi-agent mesh communication; secure emergency services APIs; patch SQL injection vulnerabilities; monitor for hallucinated vulnerability reports; detect AWVS signatures

MITRE ATLAS

AI-specific catalog annotations based on the reported behavior; these are not MITRE case-study assignments.

AML.T0040AI Model Inference API Access

12-agent AI mesh autonomously conducted offensive operations including reconnaissance, vulnerability scanning, and exploitation across multiple countries Source

Indicators of compromise

domains

  • vps.xs1.ccwu.cc
  • xss.xs1.ccwu.cc
  • c.xs1.ccwu.cc
  • attacker-collector.com
  • boss.jytian.cc.cd
  • gdp.gasjy.kdns.fr
  • gds.gasjy.kdns.fr
  • 888.bossjy.kdns.fr
  • ssh.gputw.ai

ips

  • 112.121.185.218
  • 156.240.109.26
  • 49.158.237.136
  • 202.160.76.153
  • 202.160.76.154
  • 202.160.76.155
  • 202.160.76.156
  • 202.160.76.157
  • 202.160.76.174

hashes

    urls

    • https://github.com/NousResearch/hermes-agent
    • https://github.com/0x4m4/hexstrike-ai
    • https://github.com/GreyDGL/PentestGPT
    • https://github.com/CyberStrikeAI/CyberStrikeAI

    Sources & research credit

    1 source

    Research by Eliad Kimhy, Santiago Pontiroli, Subhajeet Singha · Acronis TRU

    1. Acronis TRU: BOSSAI - Inside the AI Attack Mesh That Targeted Taiwan's Emergency Services www.acronis.com

    Acronis TRU analysis. Attribution via OPSEC failures: GitHub account TIAN0517 with email linking to 99 Gas Store (九九瓦斯行) in Hualien County; lstjks SSH key identity matching GitHub; Jy技術團隊 logo across infrastructure; home IP exposed in SSH sessions; legitimate SaaS 行銷快手 (SnapPost) Facebook auto-post tool, INCELLDERM K-beauty reseller, and gas delivery business ran from same residential IP; WeChat QR tied to Tencent iLink enterprise registration; hardcoded credentials throughout (Shodan API, Cloudflare API, MiniMax API, NVIDIA API keys)

    Record & downloads
    Source record name
    BOSSAI 12-agent AI attack mesh targeting Taiwan emergency services
    Entry ID
    bossai-12-agent-attack-mesh-2026-11
    Catalog review
    November 5, 2026
    Category
    adversary-use-of-ai
    AI involvement confidence
    confirmed