本快速入门演示如何使用 LangGraph 图 API 或函数式 API 构建计算器智能体。
  • 如果您更喜欢将智能体定义为节点和边的图，请使用图 API。
  • 如果您更喜欢将智能体定义为单个函数，请使用函数式 API。
有关概念信息，请参阅图 API 概述和函数式 API 概述。
对于此示例，您需要设置一个 Claude (Anthropic) 账户并获取 API 密钥。然后，在终端中设置 ANTHROPIC_API_KEY 环境变量。
  • 使用图 API
  • Use the Functional API

​
1. 定义工具和模型

在此示例中，我们将使用 Claude Sonnet 4.5 模型并定义加法、乘法和除法工具。
import { ChatAnthropic } from "@langchain/anthropic";
import { tool } from "@langchain/core/tools";
import * as z from "zod";

const model = new ChatAnthropic({
  model: "claude-sonnet-4-5-20250929",
  temperature: 0,
});

// Define tools
const add = tool(({ a, b }) => a + b, {
  name: "add",
  description: "Add two numbers",
  schema: z.object({
    a: z.number().describe("First number"),
    b: z.number().describe("Second number"),
  }),
});

const multiply = tool(({ a, b }) => a * b, {
  name: "multiply",
  description: "Multiply two numbers",
  schema: z.object({
    a: z.number().describe("First number"),
    b: z.number().describe("Second number"),
  }),
});

const divide = tool(({ a, b }) => a / b, {
  name: "divide",
  description: "Divide two numbers",
  schema: z.object({
    a: z.number().describe("First number"),
    b: z.number().describe("Second number"),
  }),
});

// Augment the LLM with tools
const toolsByName = {
  [add.name]: add,
  [multiply.name]: multiply,
  [divide.name]: divide,
};
const tools = Object.values(toolsByName);
const modelWithTools = model.bindTools(tools);

​
2. Define state

图的状态用于存储消息和 LLM 调用次数。
LangGraph 中的状态在智能体的整个执行过程中持续存在。带有 operator.add 的 Annotated 类型确保新消息附加到现有列表而不是替换它。
import { StateGraph, START, END } from "@langchain/langgraph";
import { MessagesZodMeta } from "@langchain/langgraph";
import { registry } from "@langchain/langgraph/zod";
import { type BaseMessage } from "@langchain/core/messages";

const MessagesState = z.object({
  messages: z
    .array(z.custom<BaseMessage>())
    .register(registry, MessagesZodMeta),
  llmCalls: z.number().optional(),
});

​
3. Define model node

The model node is used to call the LLM and decide whether to call a tool or not.
import { SystemMessage } from "@langchain/core/messages";
async function llmCall(state: z.infer<typeof MessagesState>) {
  return {
    messages: await modelWithTools.invoke([
      new SystemMessage(
        "You are a helpful assistant tasked with performing arithmetic on a set of inputs."
      ),
      ...state.messages,
    ]),
    llmCalls: (state.llmCalls ?? 0) + 1,
  };
}

​
4. Define tool node

The tool node is used to call the tools and return the results.
import { isAIMessage, ToolMessage } from "@langchain/core/messages";
async function toolNode(state: z.infer<typeof MessagesState>) {
  const lastMessage = state.messages.at(-1);

  if (lastMessage == null || !isAIMessage(lastMessage)) {
    return { messages: [] };
  }

  const result: ToolMessage[] = [];
  for (const toolCall of lastMessage.tool_calls ?? []) {
    const tool = toolsByName[toolCall.name];
    const observation = await tool.invoke(toolCall);
    result.push(observation);
  }

  return { messages: result };
}

​
5. Define end logic

条件边函数用于根据 LLM 是否进行工具调用来路由到工具节点或结束。
async function shouldContinue(state: z.infer<typeof MessagesState>) {
  const lastMessage = state.messages.at(-1);
  if (lastMessage == null || !isAIMessage(lastMessage)) return END;

  // If the LLM makes a tool call, then perform an action
  if (lastMessage.tool_calls?.length) {
    return "toolNode";
  }

  // Otherwise, we stop (reply to the user)
  return END;
}

​
6. Build and compile the agent

智能体使用 StateGraph 类构建，并使用 @[compile][StateGraph.compile] 方法编译。
const agent = new StateGraph(MessagesState)
  .addNode("llmCall", llmCall)
  .addNode("toolNode", toolNode)
  .addEdge(START, "llmCall")
  .addConditionalEdges("llmCall", shouldContinue, ["toolNode", END])
  .addEdge("toolNode", "llmCall")
  .compile();

// Invoke
import { HumanMessage } from "@langchain/core/messages";
const result = await agent.invoke({
  messages: [new HumanMessage("Add 3 and 4.")],
});

for (const message of result.messages) {
  console.log(`[${message.getType()}]: ${message.text}`);
}
To learn how to trace your agent with LangSmith, see the LangSmith documentation.
恭喜！您已经使用 LangGraph 图 API 构建了第一个智能体。
// Step 1: Define tools and model

import { ChatAnthropic } from "@langchain/anthropic";
import { tool } from "@langchain/core/tools";
import * as z from "zod";

const model = new ChatAnthropic({
  model: "claude-sonnet-4-5-20250929",
  temperature: 0,
});

// Define tools
const add = tool(({ a, b }) => a + b, {
  name: "add",
  description: "Add two numbers",
  schema: z.object({
    a: z.number().describe("First number"),
    b: z.number().describe("Second number"),
  }),
});

const multiply = tool(({ a, b }) => a * b, {
  name: "multiply",
  description: "Multiply two numbers",
  schema: z.object({
    a: z.number().describe("First number"),
    b: z.number().describe("Second number"),
  }),
});

const divide = tool(({ a, b }) => a / b, {
  name: "divide",
  description: "Divide two numbers",
  schema: z.object({
    a: z.number().describe("First number"),
    b: z.number().describe("Second number"),
  }),
});

// Augment the LLM with tools
const toolsByName = {
  [add.name]: add,
  [multiply.name]: multiply,
  [divide.name]: divide,
};
const tools = Object.values(toolsByName);
const modelWithTools = model.bindTools(tools);

// Step 2: Define state

import { StateGraph, START, END } from "@langchain/langgraph";
import { MessagesZodMeta } from "@langchain/langgraph";
import { registry } from "@langchain/langgraph/zod";
import { type BaseMessage } from "@langchain/core/messages";

const MessagesState = z.object({
  messages: z
    .array(z.custom<BaseMessage>())
    .register(registry, MessagesZodMeta),
  llmCalls: z.number().optional(),
});

// Step 3: Define model node

import { SystemMessage } from "@langchain/core/messages";
async function llmCall(state: z.infer<typeof MessagesState>) {
  return {
    messages: await modelWithTools.invoke([
      new SystemMessage(
        "You are a helpful assistant tasked with performing arithmetic on a set of inputs."
      ),
      ...state.messages,
    ]),
    llmCalls: (state.llmCalls ?? 0) + 1,
  };
}

// Step 4: Define tool node

import { isAIMessage, ToolMessage } from "@langchain/core/messages";
async function toolNode(state: z.infer<typeof MessagesState>) {
  const lastMessage = state.messages.at(-1);

  if (lastMessage == null || !isAIMessage(lastMessage)) {
    return { messages: [] };
  }

  const result: ToolMessage[] = [];
  for (const toolCall of lastMessage.tool_calls ?? []) {
    const tool = toolsByName[toolCall.name];
    const observation = await tool.invoke(toolCall);
    result.push(observation);
  }

  return { messages: result };
}

// Step 5: Define logic to determine whether to end

async function shouldContinue(state: z.infer<typeof MessagesState>) {
  const lastMessage = state.messages.at(-1);
  if (lastMessage == null || !isAIMessage(lastMessage)) return END;

  // If the LLM makes a tool call, then perform an action
  if (lastMessage.tool_calls?.length) {
    return "toolNode";
  }

  // Otherwise, we stop (reply to the user)
  return END;
}

// Step 6: Build and compile the agent

const agent = new StateGraph(MessagesState)
  .addNode("llmCall", llmCall)
  .addNode("toolNode", toolNode)
  .addEdge(START, "llmCall")
  .addConditionalEdges("llmCall", shouldContinue, ["toolNode", END])
  .addEdge("toolNode", "llmCall")
  .compile();

// Invoke
import { HumanMessage } from "@langchain/core/messages";
const result = await agent.invoke({
  messages: [new HumanMessage("Add 3 and 4.")],
});

for (const message of result.messages) {
  console.log(`[${message.getType()}]: ${message.text}`);
}
Edit the source of this page on GitHub.
Connect these docs programmatically to Claude, VSCode, and more via MCP for real-time answers.