Real-world, high-profile AI agent harnesses often rely on vendor-proprietary or opaque designs for context assembly, leaving the sources and underlying logic of assembled context poorly understood and the resulting security risks largely unexplored. In this paper, we present the first systematic analysis of context ass...
Zi-Chuan Li, Jian Cui, Ashley J. Chen et al.· 0 citations
This work presents SCRIPTIOC-BENCH, a benchmark for measuring static IOC extraction capability on real-world malicious scripts, and evaluates a broad range of proprietary and open-weight LLMs, showing that IOC recovery without execution remains challenging across model scales.
Hanna Kim, Jian Cui, Minkyoo Song et al.· 0 citations
AI agents are increasingly being developed and deployed across organizations using heterogeneous agent-development frameworks, AI models, tool interfaces, protocols, and execution environments. To realize their potential social and business impact, these agents must be able to interoperate through a common communicatio...
