Bacillus anthracis is a spore-forming pathogen that both occurs naturally in the wild and has been used for biological warfare. The detection of B. anthracis in metagenomic datasets can be confounded by the presence of other species from the genus Bacillus, which are closely related and often found in similar environmental conditions. B. anthracis encodes its pathogenicity via two plasmids, pXO1 and pXO2, and is monophyletic. Given the paramount importance of high-confidence detection for this pathogen, we constructed a targeted genomic detection panel for use with both isolates and metagenomic samples. The B. anthracis detection panel is executed on the One Codex platform, quantifying the proportion of species-specific markers detected on the core chromosome, each virulence plasmid, as well as the presence of a canonical chromosomal SNP. Tested against a collection of dozens of public Bacillus isolates, this panel clearly discriminates between B. anthracis and other closely related Bacillus species. Furthermore, an in silico limit of detection study demonstrates this high-confidence detection at as low as 0.1X sequencing depth. This targeted gene panel provides a robust tool for the high-confidence detection of B. anthracis in metagenomic samples, while also demonstrating how the One Codex platform integrates orthogonal bioinformatic methods in order to provide a comprehensive understanding of complex microbial samples.