FAIR LS ARK is a structured benchmark (FAIRsharing record, Gitbook documentation) for assessing the FAIR-enabling qualities of life science databases registered in FAIRsharing. It evaluates database-level characteristics, not individual datasets or other records contained within a database, and it uses curated FAIRsharing metadata together with linked governance artefacts as evidence. These characteristics have been chosen by evaluating and mapping a variety of these external governance frameworks for databases as well as the FAIRsharing schema; full details are available on our gitbook pages. This benchmark is intended for database managers, curators and others responsible for describing a database in FAIRsharing. It provides a common framework for interpreting FAIR in the context of database records, while allowing the same conceptual structure to be specialised for particular communities where needed. The benchmark records each metric as a transparent, citable component that can be reused by assessment tools. The metrics used within this benchmark are registered with FAIRsharing and are suitable for any database registry storing life science database records. FAIRsharing-specific operationalisations are used as example implementations within the metric and described in more detail within this document. This document was created using a template found at https://doi.org/10.5281/zenodo.17901310. This template can be copied and used within any community to begin preparations for a FAIR benchmark specific to your needs.