Abstract Aims Trial evidence indicates that glucagon-like peptide-1 receptor agonists (GLP-1 RAs) may reduce the risk of cardiovascular (CV) events in patients with diabetes and myocardial infarction (MI). We aimed to expand this observation to routine care settings. Methods and results Prospective observational study including all patients with diabetes surviving an MI and registered in the nationwide SWEDEHEART registry during 2010–17. Multivariable Cox regression analyses were used to estimate the association between GLP-1 RAs use and the study outcome, which was a composite of stroke, heart failure, Re-infarction, or CV death. Covariates included demographics, comorbidities, presentation at admission, and use of secondary CV prevention therapies. In total, 17 868 patients with diabetes were discharged alive after a first event of MI. Their median age was 71 years, 36% were women and their median estimated glomerular filtration rate was 75 mL/min/1.73m2. Of those, 365 (2%) were using GLP-1 RAs. During median 3 years of follow-up, 7005 patients experienced the primary composite outcome. Compared with standard of diabetes care, use of GLP-1 RAs was associated with a lower event risk [adjusted hazard ratio (HR) 0.72; 95% confidence interval (CI): 0.56–0.92], mainly attributed to a lower rate of re-infarction and stroke. Results were similar after propensity score matching or when compared with users of sulfonylurea. There was no suggestion of heterogeneity across subgroups of age, sex, chronic kidney disease, and STEMI. Conclusion GLP-1 RAs use, compared with standard of diabetes care, was associated with lower risk for major CV events in healthcare-managed survivors of an MI.