Seismic monitoring is essential for understanding subsurface processes, particularly in geothermal operations where low-magnitude events can provide valuable insights into reservoir behaviour. There are two significant challenges when monitoring the seismicity in Dutch geothermal operations: (1) detecting signals from seismic events as noise levels are typically high in regions hosting geothermal operations, and (2) accurately estimating their corresponding hypocentre and uncertainty. In this study, we present a comprehensive workflow for detecting and characterising low-magnitude seismic events. Specifically, we integrated data preparation, template-matching and machine-learning-based event detection, and probabilistic hypocentre estimation. Applying this workflow to 4 months of recordings in Kwintsheul, Netherlands, we detected 65 events with coherent signals, including six weak seismic events (ML < 0.0) near a local fault and a geothermal injection well. These events suggest the presence of a recurring microseismic sequence previously unreported in the area. However, spatial uncertainties, the short monitoring period, and the limited azimuthal coverage make the nature of these events unclear. Our findings highlight the importance of improving network design and refining velocity models to reduce uncertainties in event locations and magnitudes. The proposed workflow offers a scalable solution for enhancing seismic monitoring, particularly in urban and geothermal settings.