The purpose of this study was to evaluate the role of prosocial behaviour against aggression in a school‐based universal intervention adapted in two different (non‐Western) countries, Colombia and Chile. Using a randomised pretest–post‐test design (and controlling for participants' gender and parents' level of education), current results highlighted different effects of a similar programme in both sites. First, the school‐based universal programme designed for promoting prosocial behaviours in the peer context obtained a positive cross‐national effect on prosocial behaviour rated by three informants (i.e. self, peer and teacher reports). In Colombia, this effect was moderated by the initial level of prosociality of the participants and their level of education. Mediational two‐wave model corroborated that the improvement on prosocial behaviours in both countries (moderated in the case of Colombia) predicted significantly lower level of physical aggression. Characteristics of the implementation considering different cultural and historical backgrounds were discussed.