AI-Driven Multi-Agent System for Real-Time Security Analysis of Software Releases

The increasing complexity of modern software development necessitates intelligent, automated security analysis frameworks that can effectively pay attention of human on high-risk software releases. This paper introduces a MultiAgent System (MAS) framework designed to enhance the security assessment process by leveraging artificial intelligence (AI) and intelligent computing for real-time release analysis. The proposed system includes static and dynamic code analysis, anomaly detection, code verification, and architecture validation within a decentralized multi-agent architecture, ensuring scalability, adaptability, and efficiency. The approach improves security decision-making, reduces false positives, and directs expert attention efficiently. These findings underscore the importance of smart computing and multi-agent automation in software security and CI/CD cybersecurity.