Technical requirements

For conversation history, we require a CloudSQL instance running in a Google Cloud project that the user or service account that executes your LangChain code is authenticated with. Look at the Google Cloud documentation for more information about setting up and connecting to CloudSQL instances [1].

# Base components for using LangChain on Google Cloud
pip install -qU langchain-google-vertexai
# LangGraph enables creation of graph-based chat agents
pip install -qU langgraph httpx
# Grandalf is a tool to visualize graphs (like LangGraph)
pip install -qU grandalf # for visualizing the graph
# Enables LangGraph to use SQLite for memory handling
pip install -qU langgraph-checkpoint-sqlite # for chatbot memory

These libraries contain the core components of LangChain, as well as the ability to handle memory in a local MySQL database.