This paper evaluates causal impacts of a large-scale agricultural extension program for smallholder women farmers on food security in Uganda through a regression discontinuity design that exploits an arbitrary distance-to-branch threshold for village program eligibility. We find eligible farmers experienced significant increases in agricultural production, savings and wage income, which lead to improved food security. Given minimal changes in the adoption of relatively expensive inputs including HYV seeds, these gains are mainly attributed to increased usage of improved cultivation methods that are relatively costless. These results highlight the role of improved basic methods in boosting agricultural productivity among poor farmers.