Habitat suitability assessment is crucial for the conservation of endangered species in the context of dramatic global climate change. To investigate changes in the distribution pattern of forest musk in China, the present study used the optimized MaxEnt model of the ENMeval language package to simulate changes in potential suitable areas and potential suitable areas for four shared socio-economic pathways of the endangered species, forest musk (Moschus berezovskii), in both the current and the future periods, and the migration trends of the four shared socioeconomic paths in the current and future periods are analyzed. The results showed that: the MaxEnt model reached the optimal solution when RM=3.4 and FC was LQH; the dominant environmental factors for the distribution of Linnaeus musculus in China were annual precipitation, slope, and cold-season average air temperature; the current suitable habitat accounted for 23.02% of the total area of the study area, which was mainly distributed around the Hengduan Mountain Range, the area around the Longmenshan to Daliangshan Mountains, and the Qinling Mountain System; Under the future climate change, the expansion area of the suitable area is larger than the loss area, and the center of mass of the suitable area has a tendency to migrate to the northeast, in which the fluctuation area of the suitable area is smaller under the SSP370 climate scenario. Critical habitat accounts for 20.04% of the current potential suitable area, which is mainly distributed in the mountain range at the junction of Yunnan-Tibet, the junction of the Zhongshan vein and Sichuan Basin, and the Qinling mountain system. This study suggests focusing on the Yarlung Tsangpo Grand Canyon and Tsatsum Tsibagou in Tibet and the fragmented habitat of forest musk in Cangshan Erhai, Yunnan, to improve the information construction and increase the protection efforts.