Phylopeptidomics, a new metaproteomics-based concept, allows for the rapid identification of the different organisms present in a sample without any prior knowledge. The approach is based on protein extraction from complex unknown samples, followed by generation and sequencing of peptides by mass spectrometry. The high-throughput “omics” dataset is then interpreted by using a in house developed software called “µOrg.ID”, which matches sequenced peptides to taxonomical data which are then deconvoluted to discriminate species and quantify them. This method has proved successful in identifying pathogens in complex biological matrices during several Biotox-Piratox French national exercises. In this work we present the performances of this approach using a routine protocol which requires 5 hours for giving an exhaustive identification of the pathogens and other organisms present in a sample. We also improved the sample preparation and optimized the software in order to be able to give an answer within 3 hours from reception of the sample at the door of the laboratory to the list of organisms present in the sample, together with the antibiotic resistance determinants detected by mass spectrometry.