Social and emotional learning (SEL) programs are increasingly embraced by the global humanitarian sector as a potential strategy for supporting refugee children's psychosocial adaptation and learning. However, little evidence is available on the effectiveness of such SEL programs in humanitarian settings. Even less is known about whether such SEL programs can be equally effective, more beneficial, or even harmful, for children who were exposed to higher levels of conflict, in the context of war and/or postmigration settings. This study evaluates the effectiveness of a universal, multicomponent, skill-targeted SEL curriculum embedded in nonformal remedial education programming, the Five-Component SEL (5CSEL), for improving 20 proximal and distal outcomes. The large-scale cluster randomized controlled trial was conducted in 57 community sites (170 classrooms) in Lebanon with 4,289 Syrian refugee children (Mage = 9.16, SD = 2.34; 50% female). It examined both the impact of 5CSEL and the impact variation by children's level of exposure to war violence, family conflict, and school victimization. The findings do not provide conclusive evidence of the effectiveness of 5CSEL. However, we find promising signs of impacts in a number of the outcomes (effect sizes = 0.06–0.18 SD), suggesting the malleability of refugee children's SEL skills across multiple domains when given sufficient, targeted, and comprehensive support. Finally, we find little evidence of variation in impact by the level of exposure to conflict experiences, supporting the arguments for the benefits of universal SEL programming. These results provide promising directions for practitioner communities to invest in further development and revision of the SEL program contents and strategies.