Crop models indicate substantial detrimental impacts on global crop production under nuclear conflict (NC) scenarios. However, crop models have mainly been developed for assessing climate change impacts, focussing on the effects of warming and drought, giving little attention to yield penalties associated with frost damage, excess soil moisture, and flooding. The climate damage mechanisms associated with cooling and excess soil moisture are particularly relevant for evaluating impacts due to NC, but current process-based crop models have limited capacity to represent the associated impacts, leading to overly optimistic impact projections. Here, we propose to improve three well-established global gridded crop models (LPJmL, pDSSAT, CLM-crop) with respect to responses to excess moisture and cold temperatures, based on the expertise from AgMIP’s Global Gridded Crop Model Intercomparison (GGCMI). Flood damage will be accounted for in post-processing based on inundation estimates from an external flood model. These results also help to put into perspective results from a partner project that generates a large ensemble of crop model projections under different NC and management scenarios. Conclusions from this work provide new and corroborated quantitative evidence for the catastrophic indirect but global effects of nuclear conflict.