Formal models for multi-agent systems (MAS) have been introduced and studied in various areas, not only in distributed systems and AI, but also in economics and the social sciences. I will focus on logical models of the central concepts of knowledge, belief, time, and action, and will give an overview of the existing logics for MAS from a knowledge representation point of view. I will classify various MAS logics according to the epistemic and the action dimension, distinguishing individual and group action and individual and group knowledge. I will highlight problematic aspects of each of the standard accounts, including frame axioms, strategy contexts and uniform strategies.