Accurate probabilistic stability analysis of slope under rainfall effects requires a comprehensive consideration of both the randomness of rainfall and the spatial variability of soil parameters. However, existing studies typically focus on only one of these factors. To address this, a method for probabilistic stability analysis of slope that simultaneously considers both factors has been proposed. The Karhunen-Loève expansion method is employed to simulate the spatial variability of soil parameters, while a bounded random cascade model is used to simulate the randomness of rainfall. The generated rainfall time series is then applied as the upper boundary condition for the seepage and stability analysis of the slope, followed by the calculation of slope failure probability using the Latin Hypercube Sampling (LHS) method. Analysis is conducted using the landslide in Juanxuan Village, Suichuan County, Ji’an City, Jiangxi Province, as a case study. The results indicate that the probability analysis results of slope stability obtained by considering both random rainfall and spatial variability of soil parameters are consistent with engineering practice. The continuous infiltration of water stored in the terraced fields at the top of the slope gradually expands the influential range of the slope’s pore water pressure of the slope, while the increase in pore water pressure gradually diminishes, ultimately approaching zero. In comparison to rainfall infiltration, the influ of rainfall redistribution on slope stability is marginal.