Lab 1: Hello Graph - LangGraph Basics¶
⏱️ Estimated completion time: 15 minutes
Overview¶
This lab provides a minimalist introduction to LangGraph, showing how to create a simple stateful graph with just a few lines of code. You'll learn the fundamental concepts of:
- Creating a state graph
- Defining node functions
- Adding conditional edges
- Running graph execution
Learning Objectives¶
By the end of this lab, you will understand: - How to create a basic LangGraph with state management - The concept of nodes and edges in graph-based agents - How to implement conditional logic in graph flows - How to execute and monitor graph execution
Prerequisites¶
- Python 3.8+
- LangGraph installed (
pip install langgraph)
Lab Code¶
#!/usr/bin/env python3
"""
Hello LangGraph - Simple State Graph Example
--------------------------------------------
This file provides a minimalist introduction to LangGraph, showing
how to create a simple stateful graph with just a few lines of code.
"""
from langgraph.graph import StateGraph
def main():
# 1. Create a graph that works with a simple dictionary state
graph = StateGraph(dict)
# 2. Define a node function that increments a counter
def increment(state):
state["count"] = state.get("count", 0) + 1
state["message"] = f"Hello Graph! Count: {state['count']}"
return state
# 3. Add the node to the graph
graph.add_node("increment", increment)
# 4. Add a self-loop to run the increment node multiple times
graph.add_conditional_edges(
"increment",
lambda state: "increment" if state.get("count", 0) < 3 else "end",
{"increment": "increment", "end": None}
)
# 5. Set the entry point
graph.set_entry_point("increment")
# 6. Compile and run
runnable = graph.compile()
print("\n--- Running the graph ---")
result = runnable.invoke({})
print(f"Final state: {result}")
print("\n--- Step-by-step execution ---")
steps = []
for step in runnable.stream({}):
steps.append(step)
print(f"Step {len(steps)}: {step}")
print("\n--- Understanding the Example ---")
print("This simple example demonstrates core LangGraph concepts:")
print("1. State: A simple dictionary that holds our counter value")
print("2. Nodes: Functions that transform the state (our increment function)")
print("3. Edges: Define flow between nodes (our self-loop)")
print("4. Conditional Logic: Decisions about which path to take (continue or stop)")
if __name__ == "__main__":
main()
How to Run¶
- Save the code above as
01_hello_graph.py - Install dependencies:
pip install langgraph - Run the script:
python 01_hello_graph.py
Expected Output¶
--- Running the graph ---
Final state: {'count': 3, 'message': 'Hello Graph! Count: 3'}
--- Step-by-step execution ---
Step 1: {'increment': {'count': 1, 'message': 'Hello Graph! Count: 1'}}
Step 2: {'increment': {'count': 2, 'message': 'Hello Graph! Count: 2'}}
Step 3: {'increment': {'count': 3, 'message': 'Hello Graph! Count: 3'}}
--- Understanding the Example ---
This simple example demonstrates core LangGraph concepts:
1. State: A simple dictionary that holds our counter value
2. Nodes: Functions that transform the state (our increment function)
3. Edges: Define flow between nodes (our self-loop)
4. Conditional Logic: Decisions about which path to take (continue or stop)
Key Concepts Explained¶
State Management¶
- LangGraph uses a state object that flows through the graph
- Each node can read and modify the state
- State persists across node executions
Nodes¶
- Nodes are functions that transform the state
- They receive the current state and return the updated state
- Nodes represent discrete steps in your agent's logic
Conditional Edges¶
- Allow dynamic routing based on state conditions
- Enable loops, branching, and complex control flow
- Make agents adaptive and responsive to changing conditions
Next Steps¶
- Try modifying the increment function to do different operations
- Experiment with different conditional logic
- Add more nodes to create a more complex graph