Skip to main content
This feature is experimental and may change in future releases.
The wrapAnthropic() function wraps an Anthropic client to enable automatic LangSmith tracing for all message completions.

​
Installation

npm install langsmith @anthropic-ai/sdk

​
Basic usage

import Anthropic from "@anthropic-ai/sdk";
import { wrapAnthropic } from "langsmith/wrappers/anthropic";

const client = wrapAnthropic(new Anthropic());

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

​
Signature

function wrapAnthropic<T extends AnthropicType>(
  anthropic: T,
  options?: Partial<RunTreeConfig>
): PatchedAnthropicClient<T>
​
anthropic
Anthropic
required
An Anthropic client instance to wrap.
​
options
Partial<RunTreeConfig>
LangSmith tracing options (same as wrapOpenAI).
​
PatchedAnthropicClient<T>
Anthropic
The wrapped Anthropic client with automatic tracing.

​
Supported methods

The wrapper automatically traces:
  • messages.create() - Message creation (streaming and non-streaming)
  • messages.stream() - Message streams with helper methods
  • beta.messages.create() - Beta message creation
  • beta.messages.stream() - Beta message streams

​
Features

​
Automatic metadata extraction

The wrapper automatically extracts:
  • Provider (“anthropic”)
  • Model name
  • Temperature
  • Max tokens
  • Stop sequences
  • Usage metadata (input/output tokens, cache read/write)

​
System message transformation

System messages are transformed into the first message for better playground editing: 
const message = await client.messages.create({
  model: "claude-sonnet-4-20250514",
  max_tokens: 1024,
  system: "You are a helpful assistant",
  messages: [{ role: "user", content: "Hello!" }],
});

// In LangSmith, this appears as:
// messages: [
//   { role: "system", content: "You are a helpful assistant" },
//   { role: "user", content: "Hello!" }
// ]

​
Streaming support

const stream = 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") {
    process.stdout.write(event.delta.text || "");
  }
}

const finalMessage = await stream.finalMessage();

​
Usage metadata

Token usage including prompt caching is automatically tracked: 
const message = await client.messages.create({
  model: "claude-sonnet-4-20250514",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Hello!" }],
});

// Usage metadata automatically includes:
// - input_tokens
// - output_tokens
// - total_tokens
// - cache_creation_input_tokens
// - cache_read_input_tokens

​
Custom metadata

Pass additional metadata via the langsmithExtra parameter: 
const message = await client.messages.create(
  {
    model: "claude-sonnet-4-20250514",
    max_tokens: 1024,
    messages: [{ role: "user", content: "Hello!" }],
  },
  {
    langsmithExtra: {
      name: "custom-chat",
      metadata: { user_id: "123" },
      tags: ["production"],
    },
  }
);

​
Complete example

import Anthropic from "@anthropic-ai/sdk";
import { wrapAnthropic } from "langsmith/wrappers/anthropic";

const client = wrapAnthropic(new Anthropic(), {
  project_name: "my-anthropic-project",
  tags: ["production"],
});

// Non-streaming
const message = await client.messages.create(
  {
    model: "claude-sonnet-4-20250514",
    max_tokens: 1024,
    system: "You are a helpful assistant.",
    messages: [
      { role: "user", content: "What is LangSmith?" },
    ],
    temperature: 0.7,
  },
  {
    langsmithExtra: {
      metadata: { user_id: "user-123" },
    },
  }
);

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

// Streaming
const stream = client.messages.stream({
  model: "claude-sonnet-4-20250514",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Count to 10" }],
});

for await (const event of stream) {
  if (event.type === "content_block_delta" && event.delta.type === "text_delta") {
    process.stdout.write(event.delta.text);
  }
}

const finalMessage = await stream.finalMessage();

​
Tool calling

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

if (message.stop_reason === "tool_use") {
  const toolUse = message.content.find(c => c.type === "tool_use");
  console.log(toolUse);
}

​
Nested tracing

import { traceable } from "langsmith/traceable";

const myChain = traceable(
  async (input: string) => {
    const message = await client.messages.create({
      model: "claude-sonnet-4-20250514",
      max_tokens: 1024,
      messages: [{ role: "user", content: input }],
    });
    
    return message.content[0].text;
  },
  { name: "my-chain", run_type: "chain" }
);

await myChain("Hello!");

​
Notes

  • The wrapper preserves all original Anthropic SDK functionality
  • Method signatures remain unchanged except for the optional langsmithExtra parameter
  • Wrapping a client multiple times will throw an error
  • All traced calls use run_type: "llm"
  • System messages are transformed for better playground compatibility

Build docs developers (and LLMs) love

Get started for freeTalk to us