Weak-lensing bias for KiDS-1000 on eRASS1 clusters showing the calibration results on the simulation snapshots as black points and the interpolated trend as blue band (1σ uncertainty). At low redshift we are limited by the uncertainty on the multiplicative shear bias, while at higher redshift our uncertainty on the cluster member contamination dominates the systematic error budget (2σ uncertainties are indicated by the grey error-bars and light-blue curves). Hydro-dynamical modelling uncertainties, with 1σ uncertainties shown in green, remain sub-dominant.