LangGraph Simple Agent Quickstart

A simple LangGraph agent quickstart template that demonstrates the basic structure of building an AI agent with LangGraph framework.

Features

• 🚀 Quick Start: Get started with LangGraph in minutes
• 🔧 Modular Design: Clean separation of concerns with dedicated modules for graph, state, tools, and prompts
• 🛠️ Tool Integration: Built-in example tools (current time getter)
• 🎯 React Agent: Uses LangGraph's prebuilt React agent pattern
• 📦 Modern Package Management: Uses uv for fast and reliable dependency management
• 🔑 Flexible API Support: Compatible with OpenAI API and DeepSeek API

Project Structure

src/
├── agent/
│   ├── __init__.py
│   ├── graph.py        # Main LangGraph graph definition
│   ├── state.py        # Agent state management
│   ├── tools.py        # Custom tools and functions
│   ├── prompts.py      # System prompts and templates
│   └── utils.py        # Utility functions for agent creation
├── langgraph.json      # LangGraph configuration
└── pyproject.toml      # Project dependencies and metadata

Prerequisites

• Python >= 3.13
• uv package manager
• OpenAI API key or DeepSeek API key

Installation

1. Clone the repository

   git clone https://github.com/hfyydd/a_simple_agent_quickstart.git
   cd a_simple_agent_quickstart

2. Install dependencies using uv

   uv sync

3. Set up environment variables

   # Copy the example environment file
   cp .env.example .env
   
   # Edit .env and add your API keys
   # For OpenAI:
   OPENAI_API_KEY=your_openai_api_key_here
   MODEL_NAME=gpt-4
   
   # For DeepSeek:
   DEEPSEEK_API_KEY=your_deepseek_api_key_here
   DEEPSEEK_BASE_URL=https://api.deepseek.com
   DEEPSEEK_MODEL=deepseek-chat

Usage

Development Mode

Run the agent in development mode with LangGraph CLI:

uv run langgraph dev

This will start the development server and you can interact with your agent through the LangGraph Studio interface.

Direct Usage

You can also use the agent directly in your Python code:

from src.agent.graph import graph

# Run the agent
result = await graph.ainvoke({
    "messages": [{"role": "user", "content": "What time is it?"}]
})

print(result["messages"][-1]["content"])

Core Components

1. State Management (state.py)

Defines the agent's state structure using TypedDict:

class State(TypedDict):
    messages: Annotated[list, add_messages]

2. Graph Definition (graph.py)

Contains the main LangGraph graph with:

• Node definitions (chatbot_node)
• Edge connections
• Graph compilation

3. Tools (tools.py)

Custom tools available to the agent:

• get_current_time: Returns current timestamp

4. Prompts (prompts.py)

System prompts and templates for agent behavior customization.

5. Utils (utils.py)

Utility functions for agent creation and configuration.

Customization

Adding New Tools

1. Define your tool in src/agent/tools.py:

@tool
def my_custom_tool(input: str) -> str:
    """Description of what this tool does"""
    # Your tool logic here
    return result

2. Import and add it to the tools list in graph.py:

from src.agent.tools import get_current_time, my_custom_tool

agent = create_agent("chatbot", llm, [get_current_time, my_custom_tool], system_prompt)

Modifying Prompts

Edit src/agent/prompts.py to customize the agent's behavior:

system_prompt = """
Your custom system prompt here.
Define the agent's role, capabilities, and behavior.
"""

Adding New Nodes

Extend the graph by adding new nodes in graph.py:

async def my_new_node(state: State):
    # Your node logic here
    return {"messages": [new_message]}

builder.add_node("my_node", my_new_node)
builder.add_edge("chatbot", "my_node")

API Configuration

The agent supports multiple LLM providers:

• OpenAI: Set OPENAI_API_KEY and MODEL_NAME
• DeepSeek: Set DEEPSEEK_API_KEY, DEEPSEEK_BASE_URL, and DEEPSEEK_MODEL
• Other OpenAI-compatible APIs: Configure base URL and API key accordingly

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

License

This project is open source and available under the MIT License.

Support

If you encounter any issues or have questions, please:

1. Check the LangGraph documentation
2. Open an issue in this repository
3. Join the LangChain community discussions

Acknowledgments

• Built with LangGraph
• Package management by uv
• Compatible with OpenAI and DeepSeek APIs