Anthropic API Fundamentals: Official Course · Lesson 2

Messages API: Format, Response Object, and Chatbot

The messages structure with role and content, the Message object (id, stop_reason, usage), prefilling Claude's response with an assistant message, few-shot prompting, and building a simple chatbot.

20 min read3 questions in quizReady prompt includedIn progress

Practical exercise

What to do after this lesson

Write a translate(word, language) function that uses Claude to translate a word and returns only the translated word without preamble. Use few-shot examples with 2-3 pairs in messages.

Task grader

Write a translate(word, language) function that uses Claude to translate a word and returns only the translated word without preamble. Use few-shot examples with 2-3 pairs in messages.

Your answer

Ready-to-use prompt

Template for this lesson

Copy and adapt to your context. Text in angle brackets should be replaced.

from anthropic import Anthropic

client = Anthropic()
conversation = []

while True:
    user_input = input("You: ")
    if user_input.lower() == "quit":
        break
    conversation.append({"role": "user", "content": user_input})
    response = client.messages.create(
        model="claude-3-haiku-20240307",
        max_tokens=500,
        messages=conversation
    )
    reply = response.content[0].text
    print(f"Claude: {reply}")
    conversation.append({"role": "assistant", "content": reply})

Prompt sandbox

Prompt

Messages Structure

Each message is a dictionary with two keys:

messages = [ {"role": "user", "content": "What flavors are used in Dr. Pepper?"} ] response = client.messages.create( model="claude-3-haiku-20240307", max_tokens=1000, messages=messages ) print(response.content[0].text)

Rules: The list starts with a user message; roles must alternate (user → assistant → user...).

The Message Response Object

# Message(id='msg_01...', content=[TextBlock(text='...', type='text')], # model='claude-3-haiku-20240307', role='assistant', # stop_reason='end_turn', usage=Usage(input_tokens=19, output_tokens=8)) print(response.content[0].text) # response text print(response.stop_reason) # 'end_turn' or 'max_tokens' print(response.usage.input_tokens, response.usage.output_tokens)

Prefilling Claude's Response

response = client.messages.create( model="claude-3-haiku-20240307", max_tokens=500, messages=[ {"role": "user", "content": "Generate a beautiful haiku"}, {"role": "assistant", "content": "calming mountain air"} # Claude will continue from here ] ) print("calming mountain air" + response.content[0].text)

Few-Shot Prompting Through Message History

response = client.messages.create( model="claude-3-haiku-20240307", max_tokens=500, messages=[ {"role": "user", "content": "Unpopular opinion: Pickles are disgusting."}, {"role": "assistant", "content": "NEGATIVE"}, {"role": "user", "content": "I love spicy pickles from @PickleCo!"}, {"role": "assistant", "content": "POSITIVE"}, {"role": "user", "content": "Seriously why would anyone eat a pickle?"}, ] ) # Claude will respond "NEGATIVE" — it understood the pattern from examples

Simple Chatbot

conversation_history = [] while True: user_input = input("User: ") if user_input.lower() == "quit": break conversation_history.append({"role": "user", "content": user_input}) response = client.messages.create( model="claude-3-haiku-20240307", messages=conversation_history, max_tokens=500 ) assistant_response = response.content[0].text print(f"Assistant: {assistant_response}") conversation_history.append({"role": "assistant", "content": assistant_response})

Report a bug