Patterns are concise, but rich in semantic, representation of data. The approaches proposed in the literature to cope with pattern management problems usually deal with a single type of knowledge artifact and mainly concern pattern extraction issues. Little emphasis has been posed in defining an overall environment to represent and efficiently manage different types of patterns. The first general approach to deal with patterns has been proposed in the context of the PANDA project [1]. In this paper, we discuss some basic requirements for pattern manipulation and retrieval, represented according to the PANDA model. The proposed languages extend previous proposals and represent the basis for the development of an efficient pattern query processor.