Evaluating tool-augmented LLMs in conversation settings - View it on GitHub