AI researchers are striving to build complex multi-agent worlds with intended applications ranging from the RoboCup robotic soccer tournaments, to interactive virtual theatre, to large-scale real-world battlefield simulations. Agent tracking - monitoring other agent’s actions and inferring their higher-level goals and intentions - is a central requirement in such worlds. While previous work has mostly focused on tracking individual agents, this paper goes beyond by focusing on agent teams. Team tracking poses the challenge of tracking a team’s joint goals and plans. Dynamic, real-time environments add to the challenge, as ambiguities have to be resolved in real-time. The central hypothesis underlying the present work is that an explicit team-oriented perspective enables effective team tracking. This hypothesis is instantiated using the model tracing technology employed in tracking individual agents. Thus, to track team activities, team models are put to service. Team models are a concrete application of the joint intentions framework and enable an agent to track team activities, regardless of the agent’s being a collaborative participant or a non-participant in the team. To facilitate real-time ambiguity resolution with team models: (i) aspects of tracking are cast as constraint satisfaction problems to exploit constraint propagation techniques; and (ii) a cost minimality criterion is applied to constrain tracking search. Empirical results from two separate tasks in real-world, dynamic environments - one collaborative and one competitive - are provided.