[How to: add chat history](https://python.langchain.com/docs/how_to/qa_chat_history_how_to/): LLM should read this page when building a conversational question-answering application, incorporating chat history and retrieval from external knowledge sources, and deciding between using chains or agents for the application logic. Discusses building chat applications with LangChain by using chains for predictable retrieval steps or agents for more dynamic reasoning. Covers setting up components like embeddings and vector stores, constructing chains with tool calls for retrieval, and assembling LangGraph agents with a ReAct executor. Provides examples for testing the applications.

