Wildfires increasingly threaten soil integrity worldwide, causing organic matter loss, altered hydrological properties, enhanced erosion susceptibility, and disrupted nutrient cycling that can persist for years to decades. Traditional field-based assessments and satellite remote sensing provide limited spatial resolution or temporal flexibility for tracking rapid post-fire soil recovery dynamics at management-relevant scales. Hyperspectral imaging mounted on unmanned aerial vehicles (UAVs) has emerged as a transformative approach for high-resolution monitoring of burn severity impacts and subsequent soil regeneration processes. This review synthesizes recent advances in drone-based hyperspectral remote sensing for post-fire soil recovery assessment, emphasizing sensor technologies, spectral indices derivation, and soil property estimation methodologies. We examine key hyperspectral platforms including Headwall Nano-Hyperspec, Specim AFX, and HySpex sensors deployed on multirotor and fixed-wing UAVs for mapping soil charring, moisture content, organic carbon changes, and erosion risk across burned landscapes. Applications of visible-near-infrared and shortwave-infrared spectral indices for burn severity classification, coupled with machine learning algorithms for predictive modeling of soil properties and vegetation regeneration trajectories, are critically evaluated through Mediterranean, boreal, and peatland case studies. Despite promising results, challenges including atmospheric correction complexities, data processing demands, and operational constraints require continued innovation. Future directions encompass enhanced spectral-spatial fusion techniques, real-time processing capabilities, and integration with climate models for adaptive post-fire ecosystem management.