Deletions represent an important yet understudied subset of mutations in protein-coding genes, being the most abundant class of indel mutations and having a diverse range of potential impacts on protein fitness, while also being intrinsically difficult to analyze computationally for the purposes of ancestral sequence reconstruction. In order to better understand deletions and their myriad roles in shaping protein form and function, it would be to the great advantage of molecular biology researchers to be able to comprehensively profile the deletional tolerance of a protein in a simple, effective, and minimally time- and resource-intensive manner. However, whereas numerous protein engineering techniques allow for the rapid construction of highly diverse substitution libraries in proteins, there are relatively few methods available for generating libraries of deletions in a similarly effective fashion. Additionally, those methods that do currently exist for this purpose invariably exhibit limitations that render them unsuitable outside of specific contexts, highly prone to generating off-target mutations, difficult to implement experimentally, or all of the above. In this work, we introduce a novel approach for building comprehensive and unbiased deletion libraries, which we have termed SABER (SpRYgestion And Blunt-End Religation), which aims to address the aforementioned shortcomings of other deletion-library construction techniques, and moreover is exceptionally fast, simple to implement, and requires minimal specialized reagents. We use this technique to profile the landscape of tolerated deletions in Staphylococcus aureus Cas9 (SaCas9) with respect to DNA binding and discover that despite its already-small size and structural divergence from SpCas9, SaCas9 exhibits a similar degree of modularity in its ability to maintain DNA-binding function even when multiple domains have been individually removed. Additionally, our findings unexpectedly highlight the evident dispensability of a large part of a domain previously found to play an important role in DNA-binding in SpCas9, and delineate the essentiality or lack thereof of specific structural elements in the PAM-interacting region of SaCas9. We proceed to utilize this information to design minimal RNA-guided effectors which retain programmable DNA-binding activity despite missing one or more domains.