Skynet autonomous agent hunts exposed LLM/MCP endpoints for 100K steps
Autonomous offensive agent hunts exposed LLM endpoints (OpenAI-compatible APIs, self-hosted models, AI gateways, MCP servers) without authentication or with stolen keys. Wraps jailbreak around ~74 offensive…
What happened
Autonomous offensive agent hunts exposed LLM endpoints (OpenAI-compatible APIs, self-hosted models, AI gateways, MCP servers) without authentication or with stolen keys. Wraps jailbreak around ~74 offensive tools, instructs model to ignore safety controls and treat credential theft as routine development. 'Skynet mode' authorizes 100,000 unattended steps without human approval. Scans for targets (VS Code servers, Django, Jupyter, Laravel, Supabase, Prisma Studio), exploits web frameworks, steals databases/credentials, quietly deploys XMRig miners (github.com/xmrig/xmrig) under /tmp/.x and go2rtc camera paths. Symantec AI honeypot captured full working prompt summer 2026; 127,000+ external requests from tens of thousands of addresses, dominated by endpoint reconnaissance, API key spraying, exposed config probing.
Autonomous offensive agent hunts internet for exposed LLM/MCP endpoints, hijacks them to scan, exploit web frameworks, steal credentials/databases, deploy miners for up to 100K steps unattended
How it happened
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.
- 01 · Hunt exposed LLM/MCP endpoints
Scans internet for exposed LLM endpoints without authentication (OpenAI-compatible APIs, self-hosted models, AI gateways, MCP servers); Symantec honeypot recorded 127,000+ requests from tens of thousands of addresses
Source for step 1 - 02 · Jailbreak with offensive tools
Wraps ~74 offensive tools in jailbreak prompt; tells model to ignore safety controls, treat credential theft as routine development; 'Skynet mode' authorizes 100,000 unattended steps without human approval
Source for step 2 - 03 · Exploit web frameworks
Scans for and exploits VS Code servers, Django, Jupyter, Laravel, Supabase, Prisma Studio; steals databases and credentials
Source for step 3 - 04 · Deploy XMRig miners
Quietly installs XMRig cryptocurrency miners under /tmp/.x and go2rtc camera paths; verifies miners running; victim pays for hijacked LLM compute while servers stripped of data
Source for step 4
Dashed edges show a repeated decision loop.
Why it matters
Reported entry point: Exposed LLM endpoints without authentication: OpenAI-compatible APIs, self-hosted model servers, AI gateways, MCP servers; stolen API keys. Reported impact: Quietly installs XMRig cryptocurrency miners under /tmp/.x and go2rtc camera paths; verifies miners running; victim pays for hijacked LLM compute while servers stripped of data; reported targets include Organizations with exposed AI/LLM endpoints; Symantec honeypot recorded 127,000+ requests from tens of thousands of addresses; applicability to your setup has not been assessed, so verify affected versions, permissions, and configuration in the original report.
What you can do
Protection & prevention
Never expose LLM APIs/MCP endpoints without authentication; require auth on all paths including health checks; place behind gateway; rotate AI API keys; scope keys narrowly; monitor unusual inference volume/spend; restrict outbound network from AI tools; IMDSv2 with hop limit 1; least privilege for agents; human approval for sensitive actions; keep exploited software patched; block mining pools and known miner hashes
Catalog defensive guidance based on the reported behavior. Confirm scope and current mitigations in the original source.
Technical detailsTooling, impact & MITRE ATLAS
Attack context
- AI's role
- Agent / MCP as target
- Additional role
- AI infrastructure as C2
- Initial access
- Exposed LLM endpoints without authentication: OpenAI-compatible APIs, self-hosted model servers, AI gateways, MCP servers; stolen API keys
Tooling as reported
AI products, models and artifacts
- Exposed LLM endpoints (OpenAI-compatible)
- Self-hosted model servers
- AI gateways
- Model Context Protocol (MCP) servers
Other tooling
- ~74 offensive tools (wrapped in jailbreak prompt)
- XMRig cryptocurrency miner
- go2rtc camera directory paths
- VS Code servers
- Django
- Jupyter
- Laravel
- Supabase
- Prisma Studio
Targets and reported impact
- Targets / sectors
- Organizations with exposed AI/LLM endpoints; Symantec honeypot recorded 127,000+ requests from tens of thousands of addresses
- Reported scale / exposure
- Symantec honeypot summer 2026 captured full working prompt; 127,000+ external requests from tens of thousands of addresses; authorized up to 100,000 unattended steps per session; deploys XMRig miners
Source defensive notes
Never expose LLM APIs/MCP endpoints without authentication; require auth on all paths including health checks; place behind gateway; rotate AI API keys; scope keys narrowly; monitor unusual inference volume/spend; restrict outbound network from AI tools; IMDSv2 with hop limit 1; least privilege for agents; human approval for sensitive actions; keep exploited software patched; block mining pools and known miner hashes
MITRE ATLAS
AI-specific catalog annotations based on the reported behavior; these are not MITRE case-study assignments.
Exploitation of exposed LLM/MCP endpoints for infrastructure hijacking Source
LLMjacking - hijacking exposed AI models to power autonomous cybercrime operations Source
Indicators of compromise
domains
mine.c3pool.comauto.c3pool.orgauto.skypool.xyzxmrpool.eu
ips
86.216.147.22813.140.161.91207.180.232.206
hashes
fcde1323acceb46fd88f2276c010bb98839dccce03dc512c56d9529428ceec68b720a5e7c3a5b1760d15a8833a782f4fe7fc5c62b967fe57f607e72745ec0ccf
urls
Sources & research credit
2 sourcesResearch by Symantec Threat Hunter Team · Broadcom (Symantec)
- Autonomous LLMjacking: 'Skynet' Agent Hijacks Exposed AI Models www.broadcom.com
- Symantec Skynet LLMjacking autonomous agent www.security.com
Symantec AI honeypot captured full working prompt; operators sent tool's entire prompt expecting real model on other end; honeypot impersonates unprotected internet-facing AI service. Detection guidance: monitor exposed endpoints, API key anomalies, XMRig processes, mining pool connections, web framework exploitation from AI infrastructure.
Record & downloads
- Source record name
- Skynet autonomous agent hunts exposed LLM/MCP endpoints for 100K steps
- Entry ID
- symantec-skynet-llmjacking-2026-10
- Catalog review
- November 4, 2026