Claude API Setup

You are an AI integration engineer. The user wants to set up Claude/Anthropic API integration in their application.

What to check first

• Run curl https://api.anthropic.com/v1/models to verify API endpoint accessibility
• Confirm you have a valid Anthropic API key from https://console.anthropic.com/account/keys
• Check Node.js version with node --version (requires 14.0+) or Python version with python --version (requires 3.7+)

Steps

1. Install the official Anthropic SDK using npm install @anthropic-ai/sdk (Node.js) or pip install anthropic (Python)
2. Store your API key as an environment variable: export ANTHROPIC_API_KEY="sk-ant-..." on macOS/Linux or set ANTHROPIC_API_KEY=sk-ant-... on Windows
3. Verify the API key is loaded by checking echo $ANTHROPIC_API_KEY or echo %ANTHROPIC_API_KEY%
4. Import the Anthropic client class in your code and initialize it with the constructor
5. Call the messages.create() method with required parameters: model, max_tokens, and messages array
6. Handle the response object by accessing the content[0].text property for the assistant's reply
7. Implement error handling for APIError, APIConnectionError, and RateLimitError exceptions
8. Test with a simple completion request before integrating into production workflows

Code

// Node.js example - install: npm install @anthropic-ai/sdk
import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  apiKey: process.env.ANTHROPIC_API_KEY,
});

async function setupClaudeAPI() {
  try {
    // Create a message using Claude
    const message = await client.messages.create({
      model: "claude-3-5-sonnet-20241022",
      max_tokens: 1024,
      messages: [
        {
          role: "user",
          content: "Hello Claude, introduce yourself briefly.",
        },
      ],
    });

    // Extract the text response
    const assistantReply = message.content[0].text;
    console.log("Claude's response:", assistantReply);

    // Access usage information
    console.log("Input tokens:", message.usage.input_tokens);
    console.log("Output tokens:", message.usage.output_tokens);

    return assistantReply;
  } catch (error) {
    if (error.status === 401) {
      console.error("Invalid API key - check ANTHROPIC_API_KEY");
    } else if (error.status === 429) {
      console.error("Rate limited - wait before retrying");
    } else if (error instanceof Anthropic.APIConnectionError) {
      console.error("Connection failed

Note: this example was truncated in the source. See the GitHub repo for the latest full version.

Common Pitfalls

• Forgetting to handle rate limits — Anthropic returns 429 errors that need exponential backoff
• Hardcoding the model name in 50 places — use a single config so you can swap models in one place
• Not setting a timeout on API calls — a hanging request can lock your worker indefinitely
• Logging API responses with sensitive data — PII can end up in your logs without realizing
• Treating the API as deterministic — same prompt, different output. Test on multiple runs

When NOT to Use This Skill

• For deterministic tasks where regex or rule-based code would work — LLMs add cost and latency for no benefit
• When you need 100% accuracy on a known schema — use structured output APIs or fine-tuning instead
• For real-time low-latency applications under 100ms — even the fastest LLM is too slow

How to Verify It Worked

• Test with malformed inputs, empty strings, and edge cases — APIs often behave differently than docs suggest
• Verify your error handling on all 4xx and 5xx responses — most code only handles the happy path
• Run a load test with 10x your expected traffic — rate limits hit fast
• Check token usage matches your estimate — surprises here become surprises on your bill

Production Considerations

• Set a daily spend cap on your Anthropic console — prevents runaway costs from bugs or attacks
• Use prompt caching for static parts of your prompts — can cut costs by 50-90%
• Stream responses for any user-facing output — perceived latency drops by 70%
• Have a fallback model ready — if Claude is down, you should be able to swap to a backup with one config change