The Agent-Based Environment for Linking Simulations (ABELS) framework is designed to allow independent and physically distributed simulations and other data resources to form a “data cloud” for the exchange of information. In particular, it uses a distributed brokering system to facilitate dynamic linkings between independently designed, autonomous participants, without requiring the use of stringent standards to which participants must conform. This article discusses the various security threats that can arise in the ABELS architecture and presents the current design and implementation of a security framework for the system.