This paper explores the application of Large Language Models (LLMs) in power systems, focusing on their potential to revolutionize control, optimization, and decision-making processes. We present a comprehensive review of current research and applications, highlighting the challenges and opportunities in this emerging field. A practical example is provided, demonstrating the implementation of an LLM agent for power system control using Python. The power system is modeled using Pandapower, while the LLM agent is based on Llama 3, executed through Ollama. Our findings suggest that LLMs can significantly enhance the efficiency and reliability of power system operations, paving the way for more intelligent and adaptive energy management systems.