The adaptation of cutting tools is the key link of machining, although a lot of work has been done to achieve tool selection for single features, relatively little research on tool scheme adaptation for multi-feature parts as a whole. In order to take into account the factors of matching, efficiency, cost, manufacturability, and environment, a comprehensive optimization method of a multi-feature tool scheme is proposed in this work. Firstly, the critical tool of a complex feature is defined, and a tool combination method with complex single features is proposed to generate an efficient tool strategy. Secondly, a tool scheme optimization model of multi-feature parts is constructed, where each machining feature is mapped to a game player in the model space, and the set of available tools is mapped to the player strategies. In addition, the classical evolutionary game algorithm is improved to adapt to the model, according to the large difference in the number and type of features of parts. Finally, a square test piece is taken as an example to verify the proposed method, and the results show that the method presented in the paper can efficiently obtain the customer-preferred tool scheme.