How does context window size affect an Agent's memory?

• A. It limits how much past context the LLM can observe at once; relevant items must be selected for inclusion.
• B. It determines the number of tools that can be loaded into memory.
• C. It fixes the memory to a single past interaction.
• D. It expands automatically with more interactions.

Answer: (A)

Context window size determines how many tokens of history the model can see at once. That means the agent’s memory within a single turn is bounded by that limit. When the conversation or stored information grows beyond what the window can hold, the system must decide which pieces are relevant to keep visible and may summarize or retrieve other parts from external memory. The window itself doesn’t automatically expand with more interactions, and it isn’t about loading more tools into memory. So the best explanation is that the context window constrains past context visible at once and requires selecting which items to include.