{"title":"Detection of Potentially Malicious Activities in CI/CD Pipelines Based on Build Agent Behavior Analysis","authors":"E. V. Zhukovskii, V. A. Bugaev, A. A. Lyrchikov","doi":"10.3103/S0146411625700841","DOIUrl":null,"url":null,"abstract":"<p>The problem of detecting potentially malicious activity in CI/CD pipelines during the build process using build agent behavior analysis is considered. The limitations of pipeline security tools related to threat detection during builds and promising approaches for detecting malicious activity are identified. A method for detecting potentially malicious activity in pipelines based on behavioral analysis using eBPF technology for build agent profiling is proposed. The accuracy of threat detection is evaluated on a dataset containing the implementation of malicious scenarios associated with the compromise of the build process. The obtained results can be used to build agent protection tools and for further research in the field of CI/CD pipeline security.</p>","PeriodicalId":46238,"journal":{"name":"AUTOMATIC CONTROL AND COMPUTER SCIENCES","volume":"59 8","pages":"1307 - 1314"},"PeriodicalIF":0.5000,"publicationDate":"2026-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"AUTOMATIC CONTROL AND COMPUTER SCIENCES","FirstCategoryId":"1085","ListUrlMain":"https://link.springer.com/article/10.3103/S0146411625700841","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"AUTOMATION & CONTROL SYSTEMS","Score":null,"Total":0}
引用次数: 0
Abstract
The problem of detecting potentially malicious activity in CI/CD pipelines during the build process using build agent behavior analysis is considered. The limitations of pipeline security tools related to threat detection during builds and promising approaches for detecting malicious activity are identified. A method for detecting potentially malicious activity in pipelines based on behavioral analysis using eBPF technology for build agent profiling is proposed. The accuracy of threat detection is evaluated on a dataset containing the implementation of malicious scenarios associated with the compromise of the build process. The obtained results can be used to build agent protection tools and for further research in the field of CI/CD pipeline security.
期刊介绍:
Automatic Control and Computer Sciences is a peer reviewed journal that publishes articles on• Control systems, cyber-physical system, real-time systems, robotics, smart sensors, embedded intelligence • Network information technologies, information security, statistical methods of data processing, distributed artificial intelligence, complex systems modeling, knowledge representation, processing and management • Signal and image processing, machine learning, machine perception, computer vision