How to Use the Claude API with Python

sations & History Management

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    system="You are a Python code reviewer. Be direct. Point out issues first, then explain why.",
    messages=[
        {"role": "user", "content": "Review this: for i in range(len(my_list)): print(my_list[i])"}
    ]
)

print(message.content[0].text)

from dotenv import load_dotenv
from anthropic import Anthropic

load_dotenv()
client = Anthropic()

history = []

def chat(message: str) -> str:
    history.append({"role": "user", "content": message})

    response = client.messages.create(
        model="claude-sonnet-4-6",
        max_tokens=1024,
        system="You are a helpful programming assistant.",
        messages=history
    )

    reply = response.content[0].text
    history.append({"role": "assistant", "content": reply})

    return reply

print(chat("What is a decorator in Python?"))
print(chat("Show me a real example."))
print(chat("How would that work in Flask?"))

Real-Time Streaming Implementation

from dotenv import load_dotenv
from anthropic import Anthropic

load_dotenv()
client = Anthropic()

with client.messages.stream(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Explain recursion simply."}
    ]
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

print()

Production-Ready Use Case & Error Handling

from dotenv import load_dotenv
from anthropic import Anthropic

load_dotenv()
client = Anthropic()

def summarize(text: str, sentences: int = 3) -> str:
    response = client.messages.create(
        model="claude-sonnet-4-6",
        max_tokens=512,
        system=f"Summarize the following text in {sentences} sentences. Return only the summary.",
        messages=[{"role": "user", "content": text}]
    )
    return response.content[0].text

article = """
The James Webb Space Telescope has captured the deepest infrared image
of the universe ever taken. The image covers a patch of sky approximately
the size of a grain of sand held at arm's length. It contains thousands
of galaxies, some of which formed less than a billion years after the
Big Bang. Scientists believe this data will reshape our understanding
of how the earliest galaxies formed and evolved.
"""

print(summarize(article, sentences=2))

from dotenv import load_dotenv
from anthropic import Anthropic, APIError, RateLimitError, APIConnectionError

load_dotenv()
client = Anthropic()

def ask(question: str) -> str:
    try:
        response = client.messages.create(
            model="claude-sonnet-4-6",
            max_tokens=1024,
            messages=[{"role": "user", "content": question}]
        )
        return response.content[0].text

    except RateLimitError:
        return "Rate limit reached. Wait a moment and try again."

    except APIConnectionErr

Pitfall Guide

1. Virtual Environment Isolation Failure: Installing packages globally leads to dependency conflicts and broken imports. Always use python -m venv venv and activate it before running pip install. The (venv) prefix in your terminal is your only visual confirmation of isolation.
2. API Key Exposure in Version Control: Committing .env files to GitHub triggers automated credential scanning bots, leading to unauthorized usage and billing spikes within hours. Always add .env to .gitignore and load credentials via python-dotenv.
3. Stateless Context Loss: Forgetting to append the assistant response to the history list breaks conversational continuity. The API is stateless; you must manually maintain the messages array with alternating user and assistant roles, or the model will answer each prompt as an isolated query.
4. Token Limit Truncation: Setting max_tokens too low (e.g., <128) causes mid-sentence cutoffs. Start with 1024 for general tasks, and monitor message.usage.output_tokens to right-size limits for cost control without sacrificing output completeness.
5. Unhandled Rate Limits & Network Failures: LLM APIs enforce strict rate limits and experience transient network issues. Wrapping calls in try/except blocks for RateLimitError and APIConnectionError with fallback logic or exponential backoff is mandatory for production stability.
6. Ignoring System Prompt Optimization: Treating the system parameter as optional severely limits model control. Use it to enforce output formats, define roles, and constrain behavior. Proper system prompting drastically reduces hallucination and eliminates the need for heavy post-processing.

Deliverables

• Claude API Integration Blueprint: A step-by-step architectural guide covering environment setup, secure credential management, stateful conversation routing, streaming UX patterns, and production error handling strategies.
• Production Readiness Checklist: Validation steps for venv activation, .gitignore configuration, token usage monitoring, error handling coverage, rate limit fallback strategies, and system prompt effectiveness testing.
• Configuration Templates: Ready-to-use .env structure, requirements.txt (anthropic, python-dotenv), and modular Python snippets for single-turn inference, multi-turn stateful chats, and real-time streaming implementations.