The aim of the research reported in this Research Communication was to apply the 5-point body condition scoring (BCS) system to dairy buffaloes and subjectively validate it by assessing the intra- and inter-assessor agreement. For this purpose, the BCS system developed for dairy cows was applied to buffaloes. A total of 230 Nili Ravi buffaloes of varying parity, lactation stages and pregnancy status were enrolled from the Buffalo Research Institute, Pattoki, Pakistan. Four observers independently assigned BCS values to each enrolled buffalo in two phases, as follows: (1) during phase I, the assessors were trained for BCS assessment using a BCS chart developed by Elanco Animal Health Ltd.; and (2) during phase II, the assessors were trained using live buffaloes for BCS assessment. Kappa statistics (kw) were used to determine the intra- and inter-assessor agreement. The results revealed that the exact overall inter- and intra-assessor agreement was moderate (kw = 0.48–0.55) and increased to substantial levels after training on live animals (kw = 0.63–0.87). Furthermore, the intra- and inter-assessor exact agreement was higher (kw = 0.57–0.58) for buffaloes tied to the mangers compared to the buffaloes standing in the loafing area (kw = 0.50). The inter-assessor agreements within 0.25 and 0.5 points were almost perfect (kw = 0.97–1.0). The current results suggested that the 5-point BCS system (using a scale from 1 to 5 with 0.25 increments) had substantial agreement for assessment and repeatability when applied to buffaloes.