Case-Based Reasoning (CBR) is a relatively new and promising technique of artificial intelligence. Using CBR, every new problem is solved by adapting the solutions of the previously successfully solved similar problems. The intention of our research is to develop a robust and general framework which supports generation of wide-range of CBR systems using different approaches. Presented framework integrates two previously developed CBR shells: \emph{CaBaGe} and \emph{CuBaGe}. \emph{CaBaGe} (Case Base Generator) is a CBR shell for generating arbitrary decision support systems where the cases and the problems are represented as a set of values of some selected, most important attributes. \emph{CuBaGe} (Curve Base Generator) is also a CBR shell in which both the problem and the previous cases are presented in the graphical manner using curves or time-series. Presented framework, which encompasses these two shells, inherits a number of advantages including: domain independence, incremental learning, platform independence, fast retrieval algorithm, generality, and robustness.