Explainability tool integration in VizSciflow Defect Identifying within the reported issues in an issue tracker is crucial for triaging issues. Machine learning models have shown promising results regarding automated issue type prediction performance. However, we need to learn more about how such models identify defects beyond our assumptions. For example, LIME and SHAP are popular techniques to explain classifiers’ predictions. Within this study, we will close this gap in our knowledge. We will validate the prediction of defect issues on a large scale and deepen our understanding of LIME and SHAP tools for explaining machine learning models within a confirmatory study based on our assumptions. We examine if the correctness of the predictions is related to explanation quality as expected if it is easier to explain why an issue is a defect than why it is not, and the general quality of the explanations. In this paper, we will show the tool integration pipeline like LIME and SHAP in a Visually guided scientific workflow management framework (VizSciFlow) that will help the researchers to reproduce their work effortlessly.