Developed two extension templates for forecasting metrics evaluation in the sktime/sktime repository, focusing on both non-hierarchical and hierarchical, multilevel metrics. The implementation followed the established metric_detection template pattern to ensure consistency and maintainability across the codebase. Using Python and leveraging data analysis and machine learning skills, the work established a reusable foundation for evaluating forecasting metrics, enabling faster experimentation and more reliable metric reporting. No major bugs were recorded during this period, as the primary emphasis was on feature delivery and code quality improvements. The contribution addressed issue #9838 and enhanced the extensibility of forecasting metric evaluation workflows.