This study aims to improve teachers' abilities in curriculum development through workshops. This research is a collaborative supervisory action research conducted to improve and improve the quality of teachers in curriculum development. The model used in this research is the research model expressed by Kemmis and Taggart which is a development of Kurt Lewin's model. This model can include several cycles and in each cycle includes stages, namely: planning; acting and observing or implementing and observing; reflecting or reflecting, and revising plans or planning revisions. The results showed that from the pre-action, cycle I, and cycle II there were differences in the mean results of teacher achievement in curriculum development. The average pre-action score was 4.8 with a standard deviation of 0.77, in the first cycle the average score was 5.45 with a standard deviation of 0.51, and in the second cycle the average score was 7.85 with a standard deviation of 0, 37. The results of the t test, the probability value or sig (2-tailed) of 0.000 <0.05, means that there is a significant difference in the ability of teachers in curriculum development between cycle I and cycle II.