Learn how message history turns separate API calls into a multi-turn conversation.

Download the runnable notebook

This complete lesson was converted from the matching notebook stored locally in the TechClub repository. Outputs are cleared; never add API keys or private data.

Goal, prerequisites, and setup

Learn the request/response cycle and how to preserve conversation state. Know Python lists, dictionaries, functions, and environment variables. Create a virtual environment, install anthropic and python-dotenv, set ANTHROPIC_API_KEY and ANTHROPIC_MODEL, and verify the selected notebook interpreter.

Run the one-question example first. Then change the follow-up message and inspect the complete messages list. Save the redacted request sequence and explain why the assistant message must be added before the next user message.

# Optional - Install dependencies
%pip install anthropic python-dotenv

# Load environment variables
from dotenv import load_dotenv

load_dotenv()

# Create an API client
import os
from anthropic import Anthropic

client = Anthropic()

model = os.getenv("ANTHROPIC_MODEL")

def add_message(messages, role, content):
    message = {"role": role, "content": content}
    messages.append(message)

def add_user_message(messages, content):
    add_message(messages, "user", content)

def add_assistant_message(messages, content):
    add_message(messages, "assistant", content)


def chat(messages):
    message = client.messages.create(
        model=model,
        max_tokens=1000,
        messages=messages,
    )

    return message.content[0].text

# Make a starting list of messages
messages = []

# Add in the initial user question of "Define quantum computing in one sentence."
add_user_message(messages, "Define quantum computing in one sentence.")

# Pass the list of message into 'chat' to get an answer
answer = chat(messages)

# Take the answer and add it as an assistant message into our list
add_assistant_message(messages, answer)

# Add in the user's follow-up question
add_user_message(messages, "Write another sentence")

# Call chat again with the list of messages to get a final answer
answer = chat(messages)

# Take the answer and add it as an assistant message into our list
add_assistant_message(messages, answer)
print(messages)

Evidence and references