To safely perform a skill in any given environment, a robot needs to learn the preconditions for the given skill. Moreover, as robots start to operate in dynamic and unstructured environments such as our homes, these precondition models will need to generalize to variable number of objects with unknown shapes and sizes. In this work, we focus on learning precondition models of manipulation skills for such unconstrained environments. Our work is motivated by the intuition that many complex manipulation tasks, with multiple objects, can be simplified by focusing on less complex pairwise object relations. We propose an object relation model that learns continuous representations for these object relations. This object relation model is trained completely in simulation, and once learned, is used by a separate precondition learning model to predict skill preconditions for real world manipulation data. We empirically validate this precondition model on three different manipulation tasks. We show that our approach leads to significant improvements in predicting preconditions for all three tasks, across objects of different shapes and sizes.