Current understanding of landslides in Nepal is primarily based on static, catchment-scale landslide inventories or centered around data from specific events, such as the 2015 Gorkha earthquakes. While static inventories provide a useful snapshot of past landslide characteristics, we cannot use them to infer how the hazard posed by landslides may evolve in future. Automated approaches to detect landslides over large areas have recently gained traction. Many of these methods use openly-accessible cloud-based platforms to leverage large volumes of satellite imagery, inferring landslide occurrence from changes in vegetation cover as measured by NDVI. These approaches can be readily used without much background knowledge. However, the performance of these algorithms across a large spatial extent remains largely untested. In this study, we apply a readily-accessible, NDVI-based landslide detection algorithm over the full extent of Nepal to assess the ability to use NDVI to detect landslides on a national scale. From our generated annual resolution national-scale landslide probability map, we then tackle questions regarding the application of NDVI-based tools to detect landslides across a topographically- and physiographically-diverse country. We discuss the benefits of using an NDVI-based automated tool as well as locations where the application is likely to be inaccurate or misleading. This study provides a comparison of NDVI-based automated mapping approaches for different regions of Nepal, which will help to guide and interpret the accuracy of future applications.