Context Rice-based cropping system is a key for food security in the Indian subcontinent. However, the productivity of such systems has declined continuously due to deterioration in soil quality. The present study was carried out to identify key soil quality indicators and to develop indices of soil quality under long-term rice-based cropping systems in the hot sub-humid eco-region (HSE) of India. Methods Surface soil samples were collected from farmers’ field from two soil orders (Vertisols and Inceptisols) and with four long-term rice-based cropping systems: (1) rice-wheat (RW); (2) rice-chickpea (RC); (3) rice-lathyrus (RL); and (4) rice-fallow (RF). A minimum data set (MDS) of 24 soils attributes were screened to identify soil quality indicators using principal components analysis (PCA), and soil quality index (SQI) was determined. Key results The MDS indicators (in descending order of sensitivity) are field capacity (FC) (35.4%) followed by pH (30.5%) > porosity (13.6%) > potentially mineralisable carbon (PMC) (11.8%) > available boron (AB) (8.7%). Soil moisture content under FC was found as the most sensitive indicator affecting soil quality and rice equivalent yields (REY). The SQI of Vertisols (0.83) was significantly higher (P < 0.05) than Inceptisols (0.73). SQI under RC (0.86) cropping systems was significantly higher than RL (0.81), RW (0.76) and RF (0.78) (P < 0.05). Conclusions The inclusion of legumes (chickpea and lathyrus) into cropping systems sustained better soil quality and productivity. Implications SQI is not only useful for assessing soil health but also provides indicators of best-bet management practices for sustaining intensive cultivation.