NoseSense: Benchmarking Tool

NoseSense is a dynamic benchmarking tool designed to evaluate, in an automated and reproducible manner, the ability of Large Language Models (LLMs) to detect test smells, which are design anti-patterns in test code that compromise the maintainability and reliability of validation suites. The tool integrates a curated oracle of 1,000 real-world Java code instances spanning 10 distinct test smell categories and enables the simultaneous evaluation of multiple AI models, both proprietary and open-source, through an asynchronous pipeline with real-time monitoring via Server-Sent Events (SSE). Results are consolidated into an interactive analytical dashboard featuring comparative charts and CSV export, providing a robust foundation for researchers and engineers to track the evolution of LLM reasoning capabilities in software quality assurance.