Information-Centric Networking (ICN) is a new networking paradigm proposing a shift of the main network abstraction from host identifiers to location-agnostic content identifiers. So far, several architectures have been proposed implementing this paradigm shift.A key feature, common to all proposed architectures, is the in-network caching capability, enabled by the location-agnostic, explicit naming of contents. This aspect, in particular, has recently received considerable attention by the research community.However, despite this wide interest, there is a shortage of publicly-available tools suitable for evaluating the performance of caching systems effectively. In fact, all available simulators or emulators are either bound to a specific architecture or cannot execute simulations at the scale required and within a reasonable time-frame.To address these issues, we present Icarus, a Python-based caching simulator for ICN. Icarus allows users to evaluate caching strategies for any ICN implementation and also provides modelling tools useful for caching research.