Anth Hello World

Anthropic Hello World

Overview

Three minimal examples covering the Claude Messages API core surfaces: basic text completion, vision (image analysis), and streaming responses.

Prerequisites

• Completed anth-install-auth setup
• Valid ANTHROPIC_API_KEY in environment
• Python 3.8+ with anthropic package or Node.js 18+ with @anthropic-ai/sdk

Instructions

Example 1: Basic Text Message (Python)

import anthropic

client = anthropic.Anthropic()

message = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Explain quantum computing in 3 sentences."}
    ]
)

# Response structure
print(message.content[0].text)       # The actual text response
print(f"ID: {message.id}")           # msg_01XFDUDYJgAACzvnptvVoYEL
print(f"Model: {message.model}")     # claude-sonnet-4-20250514
print(f"Stop: {message.stop_reason}")# end_turn
print(f"Usage: {message.usage.input_tokens}in / {message.usage.output_tokens}out")

Example 2: Vision — Analyze an Image (TypeScript)

import Anthropic from '@anthropic-ai/sdk';
import * as fs from 'fs';

const client = new Anthropic();

// From file (base64)
const imageData = fs.readFileSync('chart.png').toString('base64');

const message = await client.messages.create({
  model: 'claude-sonnet-4-20250514',
  max_tokens: 1024,
  messages: [{
    role: 'user',
    content: [
      {
        type: 'image',
        source: {
          type: 'base64',
          media_type: 'image/png',
          data: imageData,
        },
      },
      { type: 'text', text: 'Describe what this chart shows.' },
    ],
  }],
});

console.log(message.content[0].type === 'text' ? message.content[0].text : '');

Example 3: Streaming Response (Python)

import anthropic

client = anthropic.Anthropic()

with client.messages.stream(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Write a haiku about APIs."}]
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

# Get final message with full metadata
final = stream.get_final_message()
print(f"\nTokens used: {final.usage.input_tokens}+{final.usage.output_tokens}")

Output

• Working code file with Claude client initialization
• Successful API response with text content
• Console output showing model response and usage metadata

Error Handling

Key API Parameters

Resources

• Messages API Reference
• Messages Examples
• Vision Guide

Next Steps

Proceed to anth-local-dev-loop for development workflow setup.