Rural territorial space security is not only an important part of the national security system but also the focus of territorial space planning and an important basis for realizing ecological civilization construction and rural revitalization. In addition to regional, comprehensive, and complex characteristics, rural space security has typical dynamic characteristics. Rural spaces in different periods have faced security challenges with different development goals. It is a novel exploration of the microgeographic scale to study territorial space security based on rural space. Therefore, this study takes the loess hilly region as the research area. Based on remote sensing data and statistical data, the rural territorial space security of 323 administrative villages in 18 townships in Lintao County of Dingxi City, Gansu Province, China, was quantitatively evaluated using the entropy method and spatial autocorrelation analysis method, and with the help of geographic detectors, factors affecting rural territorial space security were identified. In addition, due to the geographical particularity of the selected study area, this study fully considers the regionality and localization of the indicators in constructing the relevant evaluation index system, which makes the evaluation results more scientific. The results are as follows: (1) The overall rural territorial space security is at a moderate level in Lintao, with the level of living space security>production space security>ecological space security. Villages with a high safety level (69.73%) are primarily distributed in the irrigated agricultural areas of the Tao River Valley, whereas villages with a low safety level (30.27%) are primarily distributed in the dry farming areas in the northern and eastern hilly and mountainous areas of the county. (2) The rural territorial space security in Lintao County has obvious spatial autocorrelation. The local spatial distribution is dominated by three types: high-high (18.52%), low-high (0.62%), and low-low (9.26%). The spatial correlation between villages in the county is mainly positive correlation type and exhibits obvious agglomeration and distribution characteristics. The negative correlation types of village and town units are less distributed and more scattered, indicating their weaker spatial heterogeneity. (3) Topography, population size, government investment, and locational connections have a significant impact on the level of rural territorial space security, and the degree of impact on different townships varies. Based on these findings, subsequent research can focus on the long-term evolution process of rural territorial space security and its driving mechanism and pay attention to the understanding and change of rural territorial space security by different subjects (countries and villagers) in different periods. In the next step, attention should also be paid to the implementation and assessment system of rural territorial space security construction, and explore the establishment of a more rational and fair territorial space security evaluation system to more effectively promote the management practice of territorial space and provide services for the sustainable development of territorial space.