DEVELOPMENT OF A MODULAR MULTI-AGENT ARCHITECTURE OF THE INFORMATION SYSTEMS SECURITY TESTING SYSTEM

Chikaev Ruslan Aleksandrovich

The article emphasizes the importance of penetration testing in the light of a growing number of cyber threats and rapid technological advancements. The use of artificial intelligence (AI) in pentesting is relevant due to the complexity of attacks, technology evolution, significant financial risks of data leaks and compliance with regulatory standards. Integration of artificial intelligence into automation testing of security of information systems significantly improves the efficiency and accuracy of the testing, provides adaptation to new challenges, and a competitive advantage in the cybersecurity market by quickly identifying and responding to emerging threats. The article substantiates the concept of distributing roles between specialized agents, each of which implements a specific stage of the attack scenario - from reconnaissance and vulnerability analysis to exploitation and consolidation in the system. A model for typifying agent actions is proposed based on the MITRE ATT&CK guide to classifying and describing cyberattacks and intrusions, which allows formalizing the behavior of system components and increasing the accuracy of results verification. The developed system demonstrates stability, scalability and practical applicability in complex secure information systems.

Keywords:Modular multi-agent system, automation of penetration testing, information security, intelligent agents, coordination of actions, MITRE ATT&CK, typing of techniques, agent architecture, efficiency evaluation, pentest, autonomy.