追踪 Anthropic 应用

Python（wrap_anthropic）和 TypeScript（wrapAnthropic）中的 Anthropic 包装方法允许你包装 Anthropic 客户端，以自动记录追踪。使用包装器可确保消息（包括工具调用和多模态内容块）在 LangSmith 中良好呈现。该包装器可与 @traceable 装饰器（Python）或 traceable 函数（TypeScript）无缝配合，因此你可以用包装器追踪 Anthropic 调用，并用装饰器或函数追踪应用的其他部分。

该包装器还支持 Claude managed agents（仅 TypeScript）。请参阅追踪 Claude managed agents。

NOTE

即使使用 wrap_anthropic 或 wrapAnthropic，也必须将 LANGSMITH_TRACING 环境变量设为 'true'，追踪才会记录到 LangSmith。这样你无需修改代码即可开启或关闭追踪。

此外，你还需要将 LANGSMITH_API_KEY 环境变量设置为你的 API key（更多信息见设置。）

如果你的 LangSmith API key 关联了多个工作区，请设置 LANGSMITH_WORKSPACE_ID 环境变量以指定使用哪个工作区。

默认情况下，追踪会记录到名为 default 的项目。要将追踪记录到其他项目，请参阅将追踪记录到指定项目。

Python

import anthropic
from langsmith import traceable
from langsmith.wrappers import wrap_anthropic

client = wrap_anthropic(anthropic.Anthropic())

@traceable(run_type="tool", name="Retrieve Context")
def my_tool(question: str) -> str:
  return "During this morning's meeting, we solved all world conflict."

@traceable(name="Chat Pipeline")
def chat_pipeline(question: str):
  context = my_tool(question)
  messages = [
      { "role": "user", "content": f"Question: {question}\nContext: {context}"}
  ]
  message = client.messages.create(
      model="claude-sonnet-4-6",
      messages=messages,
      max_tokens=1024,
      system="You are a helpful assistant. Please respond to the user's request only based on the given context."
  )
  return message

chat_pipeline("Can you summarize this morning's meetings?")

TypeScript

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

const client = wrapAnthropic(new Anthropic());

const myTool = traceable(async (question: string) => {
  return "During this morning's meeting, we solved all world conflict.";
}, { name: "Retrieve Context", run_type: "tool" });

const chatPipeline = traceable(async (question: string) => {
  const context = await myTool(question);
  const messages = [
      { role: "user", content: `Question: ${question}\nContext: ${context}` }
  ];
  const message = await client.messages.create({
      model: "claude-sonnet-4-6",
      messages: messages,
      max_tokens: 1024,
      system: "You are a helpful assistant. Please respond to the user's request only based on the given context."
  });
  return message;
}, { name: "Chat Pipeline" });

await chatPipeline("Can you summarize this morning's meetings?");

追踪 Claude managed agents

wrapAnthropic 包装器还支持 Claude managed agents（仅 TypeScript）。用 wrapAnthropic 包装 Anthropic 客户端后，包装器会自动追踪代理创建、会话创建以及流经该会话的所有事件。

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

const anthropic = wrapAnthropic(new Anthropic());

// Create a managed agent
const agent = await anthropic.beta.agents.create({
  name: "my-agent",
  model: "claude-opus-4-8",
  system: "You are a helpful assistant.",
  tools: [
    // ... your tools here
  ],
});

// Create a cloud environment for the agent to run in
const environment = await anthropic.beta.environments.create({
  name: "my-environment",
  config: {
    type: "cloud",
    networking: { type: "unrestricted" },
  },
});

// Create a session connecting the agent and environment
const session = await anthropic.beta.sessions.create({
  agent: agent.id,
  environment_id: environment.id,
  title: "My session",
});

// Stream session events
const stream = await anthropic.beta.sessions.events.stream(session.id);

// Send a message to the agent
await anthropic.beta.sessions.events.send(session.id, {
  events: [
    {
      type: "user.message",
      content: [
        {
          type: "text",
          text: "Hello! Can you help me with something?",
        },
      ],
    },
  ],
});

// Consume the event stream until the session is idle
for await (const event of stream) {
  if (event.type === "session.status_idle") {
    break;
  }
}

NOTE

在 Anthropic 的多代理架构中，完整追踪子代理需要 接入单独的事件流，目前尚不支持。只有顶层 会话事件会被追踪。

---