Anthropic

This page shows how to capture Anthropic API traffic with Moda. Once the SDK is initialized, calls made with the official anthropic (Python) or @anthropic-ai/sdk (Node.js) client are captured automatically, including streaming, tool use, and extended thinking.

Prerequisites

A Moda API key (created in the dashboard at Settings → Ingestion keys; shown once at creation)
An Anthropic API key
Python 3.10 or later, or Node.js 18 or later
Node.js only: @anthropic-ai/sdk version 0.18.0 or later

Setup

Install the packages

pip install moda-ai anthropic

npm install moda-ai @anthropic-ai/sdk

Initialize Moda before creating the client

Initialize the SDK first, then construct the Anthropic client. Flush before your process exits so buffered data is sent.

import moda
from anthropic import Anthropic

moda.init("YOUR_MODA_API_KEY")

# Reads ANTHROPIC_API_KEY from the environment
client = Anthropic()

moda.conversation_id = "session_123"

response = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.content[0].text)

moda.flush()

import { Moda } from 'moda-ai';
import Anthropic from '@anthropic-ai/sdk';

await Moda.init('YOUR_MODA_API_KEY');

// Reads ANTHROPIC_API_KEY from the environment
const client = new Anthropic();

Moda.conversationId = 'session_123';

const response = await client.messages.create({
  model: 'claude-sonnet-4-20250514',
  max_tokens: 1024,
  messages: [{ role: 'user', content: 'Hello!' }],
});
console.log(response.content[0]);

await Moda.flush();

In Node.js, Moda.init() is async and the API key argument is required (it is not read from MODA_API_KEY). If Moda.init() is not awaited, calls made before initialization completes are not captured. In Python, moda.init() is synchronous and also reads the MODA_API_KEY environment variable, so moda.init() with no arguments works when that variable is set.

Verify in the dashboard

Run the script, then open Traces in the dashboard. You should see a trace with ID session_123 containing the user message and the assistant reply, with model claude-sonnet-4-20250514 and token counts. Analysis signals typically appear within minutes of ingest.

Trace grouping

Set a trace ID — the conversation_id field — before the first call of a session or run, from your session, thread, job, or run ID, so every call lands in one trace. Use moda.conversation_id (Python) or Moda.conversationId / Moda.withConversationId() (Node.js). See the Python SDK and the Node.js SDK for the full context APIs.

Streaming

Both messages.stream() and messages.create(..., stream=True) are captured. The SDK records the complete response after the stream finishes, so consume the entire stream.

import moda
from anthropic import Anthropic

moda.init("YOUR_MODA_API_KEY")
client = Anthropic()

with client.messages.stream(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Count to 5"}],
) as stream:
    for text in stream.text_stream:
        print(text, end="")

moda.flush()

import { Moda } from 'moda-ai';
import Anthropic from '@anthropic-ai/sdk';

await Moda.init('YOUR_MODA_API_KEY');
const client = new Anthropic();

const stream = await client.messages.stream({
  model: 'claude-sonnet-4-20250514',
  max_tokens: 1024,
  messages: [{ role: 'user', content: 'Count to 5' }],
});
for await (const event of stream) {
  if (event.type === 'content_block_delta' && event.delta.type === 'text_delta') {
    process.stdout.write(event.delta.text);
  }
}

await Moda.flush();

Tool use

Tool calls are captured automatically: tool_use blocks in assistant messages, and the tool_result blocks you send back in follow-up messages.

import moda
from anthropic import Anthropic

moda.init("YOUR_MODA_API_KEY")
client = Anthropic()

response = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    tools=[{
        "name": "get_weather",
        "description": "Get the current weather for a location",
        "input_schema": {
            "type": "object",
            "properties": {"location": {"type": "string"}},
            "required": ["location"],
        },
    }],
    messages=[{"role": "user", "content": "What's the weather in London?"}],
)
tool_use = next(block for block in response.content if block.type == "tool_use")
print(tool_use.name, tool_use.input)

moda.flush()

import { Moda } from 'moda-ai';
import Anthropic from '@anthropic-ai/sdk';

await Moda.init('YOUR_MODA_API_KEY');
const client = new Anthropic();

const response = await client.messages.create({
  model: 'claude-sonnet-4-20250514',
  max_tokens: 1024,
  tools: [{
    name: 'get_weather',
    description: 'Get the current weather for a location',
    input_schema: {
      type: 'object',
      properties: { location: { type: 'string' } },
      required: ['location'],
    },
  }],
  messages: [{ role: 'user', content: "What's the weather in London?" }],
});
const toolUse = response.content.find((block) => block.type === 'tool_use');
console.log(toolUse);

await Moda.flush();

Extended thinking

When extended thinking is enabled, thinking content blocks are captured alongside the final text response, in both non-streaming and streaming calls.

import moda
from anthropic import Anthropic

moda.init("YOUR_MODA_API_KEY")
client = Anthropic()

response = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=8000,
    thinking={"type": "enabled", "budget_tokens": 1024},
    messages=[{"role": "user", "content": "What is 27 * 453?"}],
)
thinking = next((b for b in response.content if b.type == "thinking"), None)
print("Has thinking block:", thinking is not None)

moda.flush()

import { Moda } from 'moda-ai';
import Anthropic from '@anthropic-ai/sdk';

await Moda.init('YOUR_MODA_API_KEY');
const client = new Anthropic();

const response = await client.messages.create({
  model: 'claude-sonnet-4-20250514',
  max_tokens: 8000,
  thinking: { type: 'enabled', budget_tokens: 1024 },
  messages: [{ role: 'user', content: 'What is 27 * 453?' }],
});
const thinking = response.content.find((b) => b.type === 'thinking');
console.log('Has thinking block:', Boolean(thinking));

await Moda.flush();

Support matrix

Capability	Python	Node.js
Messages (`messages.create`)	Supported	Supported
Streaming (`messages.stream`, `stream: true`)	Supported	Supported
Tool use	Supported	Supported
Extended thinking	Supported	Supported
System prompts	Supported	Supported
Token usage (input, output)	Supported	Supported
Prompt cache token counts (cache read / creation)	Supported	Not captured
Legacy Text Completions (`completions.create`)	Supported	Not supported
Claude on Amazon Bedrock (`AnthropicBedrock` client)	Supported	Not supported — see Amazon Bedrock

Troubleshooting

No traces appear in the dashboard. Initialize Moda before constructing the Anthropic client, and call moda.flush() (Python) or await Moda.flush() (Node.js) before the process exits. In Node.js, confirm Moda.init() is awaited and the API key starts with moda_sk_. Streaming calls show no assistant message. The response is recorded only after the stream ends. Consume the stream fully; a stream that is abandoned partway is not captured completely. Thinking blocks are missing. Extended thinking requires a model that supports it and the thinking parameter with type: "enabled" and a budget_tokens value. Every request creates a new trace. Set a trace ID before the first call of the session (see Trace grouping above).

Next steps

Python SDK — init options, context managers, and instrument controls.
Node.js SDK — init options, scoped context, and manual tracing.
Amazon Bedrock — capturing Claude models served through Bedrock.
Claude Agent SDK — capturing agents built on the Claude Agent SDK (Python).

Send data

SDKs

Providers

Frameworks and agents

OpenTelemetry

API reference

Prerequisites

Setup

Trace grouping

Streaming

Tool use

Extended thinking

Support matrix

Troubleshooting

Next steps

​Prerequisites

​Setup

​Trace grouping

​Streaming

​Tool use

​Extended thinking

​Support matrix

​Troubleshooting

​Next steps

Prerequisites

Setup

Trace grouping

Streaming

Tool use

Extended thinking

Support matrix

Troubleshooting

Next steps