The OSCARR methodology (One-pot Simple methodology for CAssette Randomization and Recombination) bridges the gap between site-directed mutagenesis and full randomization by making use of carefully designed mutagenic cassettes and an optimized one-pot megaprimer PCR. The method is especially suited to construct libraries of up to ten randomized codons for focused directed evolution, exhibits up to 97 % efficiency in the amplification of mutated over wild-type products, and is sufficiently versatile to allow mutagenesis and recombination of several cassettes within the same gene.