This study introduces XAI-CROP, an eXplainable Artificial Intelligence (XAI) algorithm designed to enhance the transparency and interpretability of crop recommendation systems (CRS) in agriculture. The algorithm leverages XAI principles to provide farmers with clear explanations for its recommendations, addressing the opacity of traditional machine learning models. XAI-CROP was evaluated against several machine learning models, including Gradient Boosting (GB), Decision Tree (DT), Random Forest (RF), Gaussian Naïve Bayes (GNB), and Multimodal Naïve Bayes (MNB), using three key performance metrics: Mean Squared Error (MSE), Mean Absolute Error (MAE), and R-squared (R2). The results demonstrated that XAI-CROP outperformed the other models, achieving a low MSE of 0.9412, an MAE of 0.9874, and a high R2 value of 0.94152, indicating its accuracy and interpretability. The algorithm uses a decision tree trained on Indian crop cultivation data and integrates the Local Interpretable Model-agnostic Explanations (LIME) technique to provide explanations for its recommendations. The study highlights the importance of XAI in agricultural decision-making, emphasizing the need for transparent and interpretable models to build trust and improve decision-making. The research contributes to the growing body of work on XAI in agriculture, offering insights into how such technologies can address challenges in food security and sustainable agriculture. The proposed XAI-CROP model is a significant advancement in crop recommendation systems, providing farmers with reliable and interpretable recommendations based on data such as soil type, weather patterns, and historical crop yields.This study introduces XAI-CROP, an eXplainable Artificial Intelligence (XAI) algorithm designed to enhance the transparency and interpretability of crop recommendation systems (CRS) in agriculture. The algorithm leverages XAI principles to provide farmers with clear explanations for its recommendations, addressing the opacity of traditional machine learning models. XAI-CROP was evaluated against several machine learning models, including Gradient Boosting (GB), Decision Tree (DT), Random Forest (RF), Gaussian Naïve Bayes (GNB), and Multimodal Naïve Bayes (MNB), using three key performance metrics: Mean Squared Error (MSE), Mean Absolute Error (MAE), and R-squared (R2). The results demonstrated that XAI-CROP outperformed the other models, achieving a low MSE of 0.9412, an MAE of 0.9874, and a high R2 value of 0.94152, indicating its accuracy and interpretability. The algorithm uses a decision tree trained on Indian crop cultivation data and integrates the Local Interpretable Model-agnostic Explanations (LIME) technique to provide explanations for its recommendations. The study highlights the importance of XAI in agricultural decision-making, emphasizing the need for transparent and interpretable models to build trust and improve decision-making. The research contributes to the growing body of work on XAI in agriculture, offering insights into how such technologies can address challenges in food security and sustainable agriculture. The proposed XAI-CROP model is a significant advancement in crop recommendation systems, providing farmers with reliable and interpretable recommendations based on data such as soil type, weather patterns, and historical crop yields.