Motivation: It is a non-trivial task to identify and design capture probes ('baits') for the diverse array of targeted-enrichment methods now available (e.g. ultra-conserved elements, anchored hybrid enrichment, RAD-capture). This often involves parsing large genomic alignments, followed by multiple steps of curating candidate genomic regions to optimize targeted information content (e.g. genetic variation) and to minimize potential probe dimerization and non-target enrichment. Results: In this context, we developed MrBait, a user-friendly, generalized software pipeline for identification, design and optimization of targeted-enrichment probes across a range of target-capture paradigms. MrBait is an open-source codebase that leverages native parallelization capabilities in Python and mitigates memory usage via a relational-database back-end. Numerous filtering methods allow comprehensive optimization of designed probes, including built-in functionality that employs BLAST, similarity-based clustering and a graph-based algorithm that 'rescues' failed probes. Availability and implementation: Complete code for MrBait is available on GitHub (https://github.com/tkchafin/mrbait), and is also available with all dependencies via one-line installation using the conda package manager. Online documentation describing installation and runtime instructions can be found at: https://mrbait.readthedocs.io. Supplementary information: Supplementary data are available at Bioinformatics online.