Natural products and their semisynthetic derivatives are an important source of drugs for the pharmaceutical industry. Bacteria are prolific producers of natural products and encode a vast diversity of natural product biosynthetic gene clusters. However, much of this diversity is inaccessible to natural product discovery. Here, we use a combination of phylogenomic analysis of the microviridin biosynthetic pathway and chemo-enzymatic synthesis of bioinformatically predicted microviridins to yield new protease inhibitors. Phylogenomic analysis demonstrated that microviridin biosynthetic gene clusters occur across the bacterial domain and encode three distinct subtypes of precursor peptides. Our analysis shed light on the evolution of microviridin biosynthesis and enabled prioritization of their chemo-enzymatic production. Targeted one-pot synthesis of four microviridins encoded by the cyanobacterium Cyanothece sp. PCC 7822 identified a set of novel and potent serine protease inhibitors, the most active of which had an IC50 value of 21.5 nM. This study advances the genome mining techniques available for natural product discovery and obviates the need to culture bacteria.