Ledger / Nov 2026
CyberStrikeAI framework automates post-exploitation against educational institutions
Confirmed
Kaspersky, which analyzed the intrusions, states the framework's LLM planned and generated the commands.
- Model
- Not namedCyberStrikeAI
- Agent harness
- CyberStrikeAIopen-source harness
- Attributed country
- Not attributedthe source names no country
- Sectors hit
- educationEducational institutions in Africa and Latin America
Kaspersky analyzed September to October intrusions at educational institutions in African and Latin American countries where attackers used the open-source CyberStrikeAI framework, whose server-side LLM plans and generates the commands its implant runs.
“The first step involves deploying the CyberStrikeAI C2 agent, which receives commands from a server-based LLM orchestrator.”Kaspersky
Sources
- CyberStrikeAI framework automates post-exploitation against educational institutionsKaspersky · Nov 2026
- CyberStrikeAI attacks educational institutionsKaspersky · Oct 2026
- Date
- Nov 2026
- Actor
- Unattributed Chinese-speaking operator (per Kaspersky) (unknown), as attributed by Kaspersky
- Nexus
- Chinese-speaking; country attribution unestablished
- Targets
- Educational institutions in Africa and Latin America
- What the AI did
- Agentic intrusion, Ops automation
- Model as named
- CyberStrikeAI
- Model basis
- A researcher names it
- Autonomy
- Semi-autonomous
- Outcome
- Domain admin was reached within three to four days of agent deployment in one case.