Acronis Threat Research Unit reconstructed BOSSAI, a mesh of 12 AI agents built by a single Taiwan-based operator and used against hundreds of organisations in 11 countries. The agents split infrastructure work from attack work: Grok, Claude Code, Codex and OpenClaw built and maintained the platform, while Hermes, Cursor, Trae and CyberStrike scanned targets and ran exploits through MCP servers including Kali, HexStrike AI and Ghidra. The operator mostly pasted scan results between agents and let Hermes work; one terse instruction to extract student data led to enumeration of more than 50 subdomains, a SQL injection against an Oracle 9i system and an attempt to dump 2,563 administrator records. Confirmed breaches at universities and emergency services providers exposed medical data and personal information. Commercial models refused the offensive work, so the operator relied on abliterated local DeepSeek, GLM and Qwen variants.
Acronis traces a lone operator's 12-agent BOSSAI mesh across 11 countries
Acronis Threat Research Unit reconstructed BOSSAI, a mesh of 12 AI agents built by a single Taiwan-based operator that was used against hundreds of organisations in 11 countries, with confirmed breaches at universities and emergency services providers exposing medical data and personal information. The agents divided labour between infrastructure β Grok, Claude Code, Codex and OpenClaw β and attack work β Hermes, Cursor, Trae and CyberStrike β scanning targets and running exploits through MCP servers including Kali, HexStrike AI and Ghidra; one terse instruction to extract student data led to enumeration of more than 50 subdomains, a SQL injection against an Oracle 9i system and an attempt to dump 2,563 administrator records. Because commercial models refused the offensive work, the operator relied on abliterated local DeepSeek, GLM and Qwen variants.
Treat MCP servers such as Kali, HexStrike AI and Ghidra as high-risk egress points and gate agent tool calls behind approval, since one terse instruction in this mesh produced 50+ subdomain enumerations and a SQL injection.
more in playbook: Securing agents and AI toolsRun your AI side-project on zahid.host
EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain β perfect for shipping the agent you just read about.