Customer Support - LangGraph Starter

A streamlined example of an AI agent built using LangGraph. This project demonstrates the absolute essentials of graph-based LLM orchestration, focusing on a single-node "agent" pattern.

🚀 Overview

Unlike traditional linear chains, this project uses a stateful graph approach to handle customer inquiries. While simple, it establishes the architecture needed for complex, multi-step agentic workflows.

Key Features

• State Management: Uses MessagesState to track conversation history.
• Graph Architecture: Implements a StateGraph with a single functional node.
• LLM Integration: Direct invocation of a Chat Model within the graph logic.

🛠️ Tech Stack

• LangGraph - Agent orchestration framework.
• LangChain - LLM utility library.
• Python 3.10+

📖 How it Works

The workflow follows a basic START ➔ Agent Node ➔ END lifecycle:

1. START: The graph receives an input message.
2. Agent Node: A function calls the LLM, passing the current state.
3. END: The LLM's response is appended to the state and returned to the user.