For research data to be reusable by scientists or machines, the data and associated meta-data should comply with the so-called "FAIR principles", i.e. it should be findable, accessible, interoperable, and reusable [1]. To realize this, is not a straightforward task, as researchers do not know how FAIR or un-fair their data actually is and how to improve their FAIRness. A quantitative measure, which is easy to apply could help. The Helmholtz Metadata Collaboration (HMC) works on improving tools to automate the assessment of the FAIRness of publications. The F-UJI framework [2] originating from the FAIRsFAIR project is a powerful tool that provides a score for the FAIRness for machine findable and readible metadata of a given publication with respect to the FAIRsFAIR metric [3]. We co-develop F-UJI to explore and evaluate ways to apply it in user-sided tools. On this poster we present a FAIR assessment through F-UJI of Helmholtz data repositories. With our work, we want to identify gaps in the Helmholtz data infrastructure with respect to the FAIRness of (meta)data and how these gaps can be closed effectively. We also provide an outlook on possible research into the development of FAIRness over time within communities. [1] Wilkinson, M.D.et al. Sci Data 3, 160018 (2016) [2] Devaraju, A., Huber, R. (2020). F-UJI, Zenodo. https://doi.org/10.5281/zenodo.4063720 [3] Devaraju, A., et al. (2020). FAIRsFAIR Metrics. Zenodo. https://doi.org/10.5281/zenodo.6461229