Microgrids that integrate high-penetration renewable energy sources (RESs), seawater desalination, energy storage, and hydrogen production are promising ways to reduce RESs curtailment and improve the system operational economics. This work investigates the holistic planning and operation of the hybrid hydrogen-electrical microgrids. To mitigate the uncertainty associated with RESs power outputs, we propose a scenario-based two-stage stochastic programming (SP) where the first stage and the second stage aim to minimize the total investment costs and the operational costs, respectively. The proposed SP optimizes the capacities of microgrids facilities while considering investment and operational costs (e.g., daily capital depreciation, RESs curtailment costs, and electricity purchasing costs). An improved scenario reduction method which incorporates Cholesky decomposition and k-means clustering algorithm is employed to solve this problem. Case studies are conducted on a microgrid test system. The results show that the proposed two-stage SP demonstrates enhanced utilization of RESs and reducing total investment and operational costs. The safe and stable operation of the hydrogen-electrical microgrids is also satisfied.