Developed a range-based task evaluation feature for the UKGovernmentBEIS/control-arena repository, enhancing the CLI to accept START-END strings for targeted task assessment. Leveraging Python and argument parsing skills, the solution parses user input, converts it into a (start, end) tuple, and integrates seamlessly with the existing configuration framework. This update streamlines the evaluation process by reducing manual configuration and enabling precise performance analysis within defined planning windows. The work also included linting improvements and addressed a previously reopened issue, reinforcing code quality and reliability. Overall, the contribution focused on CLI development to improve workflow efficiency and task-range analysis capabilities.