Building Your First AI App with the Claude API

Claude Code is for the terminal, but if you need AI capability inside your own app (a website, a backend service, a chatbot), you use the Anthropic API directly.

Setup

pip install anthropic

Generate an API key from the Anthropic Console and save it as an environment variable:

export ANTHROPIC_API_KEY="your-api-key-here"

Your First API Call

import anthropic

client = anthropic.Anthropic()  # picks up the API key from the env variable automatically

message = client.messages.create(
    model="claude-sonnet-4-5",  # check your Anthropic dashboard for the latest model ID
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Write a 3-line poem about the mountains of Nepal."}
    ]
)

print(message.content[0].text)

The messages array is the conversation history — you have to send the full context every time, since the API is stateless (there's no built-in memory between requests).

Controlling Behavior with a System Prompt

message = client.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=1024,
    system="You are a friendly customer support assistant who only replies in short, clear sentences.",
    messages=[
        {"role": "user", "content": "When will my order arrive?"}
    ]
)

The system parameter sets the model's "personality" and rules — it applies to every turn of the conversation.

Just as Simple in JavaScript

import Anthropic from '@anthropic-ai/sdk';

const client = new Anthropic();

const message = await client.messages.create({
  model: 'claude-sonnet-4-5',
  max_tokens: 1024,
  messages: [{ role: 'user', content: 'Hello Claude!' }],
});

console.log(message.content[0].text);

Key Takeaways

• The API is stateless — the full conversation history has to be sent with every request.
• The system parameter controls the model's behavior/personality.
• Official SDKs exist for both Python and JavaScript, and integration is very similar in both.