Our national biodiversity collections are massive repositories of biological specimens, such as plants, fungi and insects, carefully curated by subject experts. Locked within the genomes of these specimens are potential targets for biodiscovery. Our aim was to develop a pipeline to skim the genomes derived from collection specimens to mine for peptide sequences with prospective biomedical applications. Here, we focused on the Pompilidae, a family of venom-producing spider wasps1. These wasps hunt spiders to lay their eggs, and their developing larva feed on the paralysed spider's body. Some of the powerful toxins that create temporary paralysis on spiders have been shown to modulate major voltage-gated sodium channels2 such as those found in the human brain, with prospective therapeutic uses in the treatment of Alzheimer's, Parkinson's, and other conditions affecting the central nervous system. Using fresh spider wasp venom and glands, we characterised toxins present through transcriptomics and mass spectrometry. These are used as references for our bioinformatics pipeline, which can then mine hundreds of de novo assembled genomes to find homologous sequences. Our pipeline can be easily adapted for other groups of interest within collections, such as lichens and velvet ants. This study shows how genomic technologies can be used to unlock the incredible potential of our biodiversity collections.