After establishing what LLM agents are, we now look at their internal structure. This chapter breaks down an LLM agent into its essential components. You will learn about:

• The Large Language Model (LLM) serving as the agent's cognitive core.
• How instructions and prompts are crafted to direct agent behavior.
• The function of tools in extending an agent's capabilities to interact with external systems.
• The concept of short-term memory for maintaining context during operations.
• Basic methods an agent uses for planning and sequencing actions.
• The cyclical process of observation, thought, and action that defines an agent's operation.

Understanding these individual parts and how they fit together is key to grasping how agents process information and accomplish goals.