Hydropower and renewable energy prediction require accurate precipitation forecast as fundamental, which is computed from Numerical Weather Forecast (NWP) models. Assimilating satellite data has become an important method to improve NWP model results, especially geostationary meteorological satellites (GMS) that can provide continuous observation information of weather systems. The Advanced Geostationary Radiance Imager (AGRI) onboard China’s new generation GMS, FY-4A, can provide observations of temperature and water vapor covering the land and surrounding sea areas. Using a newly established WRF-GSI NWP system, this paper analyzes the impact of AGRI data assimilation on several precipitation case forecasts in China. First, channel selection of AGRI data, along with data thinning and observation error adjustment are performed. Then, four experiments of precipitation case forecasts are conducted and compared with each other. The results show that the influence of assimilating conventional data is mainly located in the inland area of China, while AGRI data can provide more observation information in the offshore ocean area. AGRI data assimilation can significantly improve the accuracy of atmospheric temperature and water vapor at a height of 500 hPa in the coastal area of the model, which further enhanced precipitation forecasts. The improvement of ETS score for heavy precipitation is most obvious, the biggest improvement can be up to 60% in the 15 July 2021 case, indicating the addition of AGRI data can ensure the improvement of the ETS scores of heavy precipitation that larger than 50 mm. The results prove that the AGRI data has added value to precipitation forecasts in the WRF-GSI NWP system, the improved forecast accuracy has great potential on hydrological forecasts, supporting the development of related hydropower applications.