Food and waterborne protozoan pathogens including Cryptosporidium parvum, Giardia enterica and Toxoplasma gondii are a global concern for human public health. While all three pathogens have been detected in commercial shellfish, there is currently no standard approach for detecting protozoan parasites in shellfish. Common molecular and microscopic methods are limited in the number of pathogens they can simultaneously detect and are often targeted at one or two of these pathogens. Previously, we developed and validated a novel 18S amplicon-based next-generation sequencing assay for simultaneous detection of Cryptosporidium spp., Giardia spp. and T. gondii in shellfish. In this study, we applied the assay for protozoan pathogen detection in wild oysters from Prince Edward Island (PEI). Oysters were harvested from restricted and prohibited areas, classified by the Canadian government according to fecal coliform counts in surrounding waters, and different fractions (whole tissue homogenate and hemolymph) were analyzed. Protozoan DNA was detected using metabarcoding in 28%, of oysters tested (N = 128), and the pathogen read counts in oyster homogenate were considerably higher than those in hemolymph. Protozoan read count thresholds were established for classifying probable oyster contamination with pathogens to account for low levels of background protozoan reads detected in negative controls. Assay results showed protozoan contamination was not associated with harvesting site classifications, suggesting that using fecal indicators for ensuring food safety may be insufficient. Due to the complex matrix, an oyster DNA reduction step may further improve the pathogen detection sensitivity of the assay. Results from this study affirm that novel metabarcoding is a promising screening tool for detection of protozoan pathogens in shellfish.