The reconstruction of metagenome-assembled genomes (MAGs) from environmental and host-associated microbiomes is a powerful technique to obtain taxonomic and functional information for uncultivated microbes. However, the multiple bioinformatic steps to obtain MAGs can be cumbersome and computationally inefficient for large data sets. Here we present Metaphor, a user-friendly workflow designed to facilitate quality control, assembly and binning of metagenomes. Metaphor incorporates best practices in Snakemake workflow development, optimizing computational efficiency and enabling large-scale data analyses. Our pipeline has flexible options for individual or pooled assembly and binning, and multiple binning algorithms with a binning refinement step. Metaphor also returns gene and contig-level coverage estimation, functional and taxonomic annotation, performance benchmarking (runtime and memory usage per computational task), and a variety of tabular and graphical reports generated by each of its six modules: quality control, assembly, annotation, binning, read mapping, and postprocessing. Final data generated by Metaphor is designed to be easily imported into downstream applications for further statistical and bioinformatic analyses. We showcase Metaphor with metagenome datasets from the CAMI challenge, an initiative for the performance evaluation of metagenomics software, and demonstrate how to fine-tune Metaphor for different datasets.