Agricultural ecological efficiency (AEE) is an important index to measure the coordinated development of agricultural production, resources, and the environment. This study tries to explore the impact of rural human capital (RHC) on AEE, hoping to provide a reference for promoting the green development of agriculture. Data sets (2006–2021) used in this study were gathered from Official Statistics Panel data in 30 provinces in China. Firstly, the super-efficient slacks-based measure model (SBM) with non-expected output is employed to effectively measure AEE. Secondly, the entropy method is used to measure RHC in three dimensions: education, health, and migration. Finally, this study discusses the impact of RHC on AEE using the panel Tobit model and further verifies the moderating effect of Internet popularization on the effect of RHC on AEE using the moderating effect model. The results show that RHC has a significant positive impact on the promotion of AEE, and this positive impact has obvious regional heterogeneity. Specifically, RHC has a greater promoting role on AEE in the eastern and northeastern regions, while it has a smaller promoting role on AEE in the central and western regions. In addition, Internet penetration plays a positive moderating role in the mechanism of the effect of RHC on the AEE. This study may serve as a reference for improving AEE by providing theoretical guidance and policy suggestions for promoting agricultural green development.

Against the backdrop of increasingly severe agricultural labor aging (ALA), the aging process not only threatens food security but also poses challenges to green and sustainable agricultural development. Existing studies have paid insufficient attention to how ALA simultaneously affects grain production efficiency (GPE) and grain eco-efficiency (GEE), and there is a particular lack of systematic investigation into the moderating roles of different crop types and social networks. To address this gap, this study utilizes survey data from 1056 farm households across five major grain-producing provinces in China and employs Tobit regression models to empirically examine the dual effects of ALA on GPE and GEE, while also revealing the moderating mechanisms of formal and informal dual-layer social networks. The main findings are as follows: (1) ALA generally inhibits both GPE and GEE across different grain crops, with a more prevalent negative impact on GEE. (2) The impact of ALA on the two types of efficiency exhibits crop-specific nonlinear characteristics: a positive U-shaped relationship for maize, an inverted U-shaped relationship for rice, and no significant nonlinear relationship for wheat. (3) Social networks play significant linear and nonlinear moderating roles in mitigating the negative effects of ALA, though their effects vary depending on network type, crop system, and efficiency dimension. Based on these findings, it is recommended to implement differentiated intervention strategies tailored to crop characteristics and aging stages, build a multi-tiered social network support system, and strengthen the research, extension, and service support for green technologies targeting middle-aged and older farmers, thereby synergistically enhancing grain production capacity and ecological sustainability.