Python Playground

An interactive Python sandbox powered by Pyodide — run Python entirely in your browser via WebAssembly.

Features

• Full Python 3.11 runtime
• Edit and run code instantly
• No setup required

Limitations

• Cannot call actual LLMs (no Ollama/OpenAI APIs)
• Cannot install system-dependent packages

---

Example 1: Basic State

Explore how LangGraph's TypedDict state works.

Basic State Management

from typing import TypedDict, Annotated from operator import add # Define a simple state schema class GraphState(TypedDict): messages: Annotated[list[str], add] counter: int user_input: str # Create initial state state = GraphState( messages=['System initialized'], counter=0, user_input='Hello' ) print('Initial State:') print(f'Messages: {state["messages"]}') print(f'Counter: {state["counter"]}') print(f'User Input: {state["user_input"]}') # Simulate state update state['messages'].append('User said: ' + state['user_input']) state['counter'] += 1 print('\nUpdated State:') print(f'Messages: {state["messages"]}') print(f'Counter: {state["counter"]}')

Key Concepts:

• TypedDict defines state structure
• Annotated[list, add] merges list updates
• State flows between nodes

---

Example 2: Graph Nodes

Build a basic graph structure with nodes.

Graph Structure

from typing import TypedDict # Define state class State(TypedDict): input: str output: str step: int # Define node functions def node_a(state: State) -> dict: print(f'Executing Node A (Step {state["step"]})') return { 'output': state['input'].upper(), 'step': state['step'] + 1 } def node_b(state: State) -> dict: print(f'Executing Node B (Step {state["step"]})') return { 'output': state['output'] + ' - processed', 'step': state['step'] + 1 } # Simulate graph execution state = {'input': 'hello langgraph', 'output': '', 'step': 1} print('Starting graph execution...\n') state = {**state, **node_a(state)} state = {**state, **node_b(state)} print(f'\nFinal Output: {state["output"]}') print(f'Total Steps: {state["step"] - 1}')

Key Concepts:

• Nodes are functions that take and return state
• State updates are merged automatically
• LangGraph manages flow between nodes

---

Example 3: Conditional Routing

Learn conditional routing based on state.

Conditional Routing

from typing import TypedDict, Literal class RouterState(TypedDict): message: str sentiment: str route: str def analyze_sentiment(state: RouterState) -> dict: message = state['message'].lower() positive = ['good', 'great', 'excellent', 'happy', 'love'] negative = ['bad', 'terrible', 'sad', 'hate', 'angry'] pos = sum(1 for w in positive if w in message) neg = sum(1 for w in negative if w in message) sentiment = 'positive' if pos > neg else 'negative' if neg > pos else 'neutral' print(f'Analyzed: "{message}" -> {sentiment}') return {'sentiment': sentiment} def route_message(state: RouterState) -> str: return state['sentiment'] # Test messages messages = ['I love this!', 'This is terrible', 'It is okay'] for msg in messages: print(f'\n{"=" * 40}') state = {'message': msg, 'sentiment': '', 'route': ''} state = {**state, **analyze_sentiment(state)} route = route_message(state) print(f'Routed to: {route.upper()} handler')

Key Concepts:

• Routing functions return next node name
• Enables dynamic flow control
• Use add_conditional_edges() in LangGraph

---

Next Steps

Ready for real LLMs? Start with:

• 01. Chatbot Basics
• 02. Tool Calling
• 03. Memory & Persistence

Having Issues?

Pyodide takes a few seconds to initialize. Refresh if needed.