With growing penetration of wind power into the power grid, while achieving low cost sustainable electricity supply, it also introduces technical challenges with the associated intermittency. This paper proposes a fully adaptive Was-serstein-based distributionally robust multi-stage frame-work based on mixed decision rules (MDR) for uncertain unit commitment problem (UUC) to better adapt wind pow-er respecting non-anticipativity both in unit state decision and dispatch process. Comparing with the existing mul-ti-stage model, the proposed framework introduces an im-proved MDR to handle all decision variables to expand feasible region, thus, this framework can obtain various typical models by adjusting the number of relevant periods of decision variables. As a result, our model can find feasi-ble solution to some problems that are not feasible in the traditional models while finding better solution to feasible problems. The proposed model is reformulated with ad-vanced optimization method and improved MDR to form the mixed integer linear programming (MILP) model to address the computational intractability. The effectiveness and efficiency of the proposed model have been validated with case studies using IEEE benchmark systems.