TrustStream: How Much Should We Trust AI Agents? Visual Analytics for Alignment Monitoring in Multi-Agent Systems

Authors

Aaron Wang (University of British Columbia), Justin Li (Carmel High School), Sizheng Cailean Chen (West Lafayette Junior Senior High School), Andralyn Yao (West Lafayette Junior Senior High School), Nabin Khanal (Purdue University), Qi Yang (Purdue University), Jasmine Tang (McGill University), Zhancheng Su (Carmel High School), Jieqiong Zhao (Augusta University), Hannah Yanhua Zong (Purdue University ), Zhenyu Cheryl Qian (Purdue University), Yingjie Victor Chen (Purdue University)

Abstract

Diagnosing alignment failures in multi-agent systems requires analysts to relate evolving agent behavior to communication pathways and scoped context. We present TrustStream, a visual analytics system combining reviewable rubric-based alignment assessments with temporal conversation graphs, per-agent trends, and communication flow views. Applied to the 2026 VAST Challenge MC-1 dataset, TrustStream revealed low-scoring private communication before a data embargo breach, identified changing violation categories, and traced how misaligned behavior spread into a system-wide breakdown. Applied to AI multi-agent systems, TrustStream allows for incident analysis and longer-term alignment monitoring. Source code available on GitHub: https://github.com/nesteagle/trust-stream