TL;DR
In recent years, large language models (LLMs) have evolved beyond text-based chatbots into agents capable of executing tools—functions that let them gather new information, interact with external systems, or even take actions that affect the real world.