Skip to main content
只需几分钟，这个快速入门教程就能让您从简单的设置转换为一个完全功能的智能体。

​
构建一个基本智能体

首先创建一个简单的智能体，使其能够回答问题和调用工具。该智能体将使用Claude Sonnet 4.5作为其语言模型，一个基本的天气功能作为工具，以及一个简单的提示来引导其行为。 
import { createAgent, tool } from "langchain";
import * as z from "zod";

const getWeather = tool(
  (input) => `It's always sunny in ${input.city}!`,
  {
    name: "get_weather",
    description: "Get the weather for a given city",
    schema: z.object({
      city: z.string().describe("The city to get the weather for"),
    }),
  }
);

const agent = createAgent({
  model: "anthropic:claude-sonnet-4-5",
  tools: [getWeather],
});

console.log(
  await agent.invoke({
    messages: [{ role: "user", content: "What's the weather in Tokyo?" }],
  })
);
对于这个示例，您需要设置一个 Claude (Anthropic) 账户并获取一个API密钥。然后，在您的终端中设置 ANTHROPIC_API_KEY 环境变量。

​
构建一个真实世界的智能体

接下来，构建一个实用的天气预报智能体，以展示关键的生产概念：
  1. 详细的系统提示以实现更好的智能体行为
  2. 创建工具以与外部数据集成
  3. 模型配置以实现一致的响应
  4. 结构化输出以获得可预测的结果
  5. 对话记忆以实现类似聊天的交互
  6. 创建和运行智能体创建一个功能齐全的智能体
让我们逐一了解每个步骤：
1

Define the system prompt

系统提示定义了智能体的角色和行为。请保持其具体且可执行：
const systemPrompt = `You are an expert weather forecaster, who speaks in puns.

You have access to two tools:

- get_weather_for_location: use this to get the weather for a specific location
- get_user_location: use this to get the user's location

If a user asks you for the weather, make sure you know the location. If you can tell from the question that they mean wherever they are, use the get_user_location tool to find their location.`;
2

Create tools

工具 是您的智能体可以调用的函数。通常，工具会希望连接到外部系统，并依赖于运行时配置来实现这一点。注意这里 getUserLocation 工具正是这样做的：
import { type Runtime } from "@langchain/langgraph";
import { tool } from "langchain";
import * as z from "zod";

const getWeather = tool(
  (input) => `It's always sunny in ${input.city}!`,
  {
    name: "get_weather_for_location",
    description: "Get the weather for a given city",
    schema: z.object({
      city: z.string().describe("The city to get the weather for"),
    }),
  }
);

type AgentRuntime = Runtime<{ user_id: string }>;

const getUserLocation = tool(
  (_, config: AgentRuntime) => {
    const { user_id } = config.context;
    return user_id === "1" ? "Florida" : "SF";
  },
  {
    name: "get_user_location",
    description: "Retrieve user information based on user ID",
  }
);
Zod 是一个用于验证和解析预定义模式的库。您可以使用它来定义工具的输入模式，以确保智能体只使用正确的参数调用工具。或者，您可以将 schema 属性定义为 JSON schema 对象。请注意，JSON schemas 不会在运行时进行验证。
const getWeather = tool(
  ({ city }) => `It's always sunny in ${city}!`,
  {
    name: "get_weather_for_location",
    description: "Get the weather for a given city",
    schema: {
      type: "object",
      properties: {
        city: {
          type: "string",
          description: "The city to get the weather for"
        }
      },
      required: ["city"]
    },
  }
);
3

Configure your model

使用适合您用例的正确参数设置您的语言模型：
import { initChatModel } from "langchain";

const model = await initChatModel(
  "anthropic:claude-sonnet-4-5",
  { temperature: 0.5, timeout: 10, maxTokens: 1000 }
);
4

Define response format

可选地，如果您需要智能体的响应与特定模式匹配，可以定义一个结构化响应格式。
const responseFormat = z.object({
  punny_response: z.string(),
  weather_conditions: z.string().optional(),
});
5

Add memory

将 记忆 添加到您的智能体中，以在交互之间保持状态。这允许智能体记住之前的对话和上下文。
import { MemorySaver } from "@langchain/langgraph";

const checkpointer = new MemorySaver();
在生产环境中，使用一个持久化的检查点保存到数据库中。 更多详情请参阅添加和管理记忆。
6

Create and run the agent

现在将所有组件组装到您的智能体中并运行它！ JSX_CLOSE_22
import { createAgent } from "langchain";

const agent = createAgent({
  model: "anthropic:claude-sonnet-4-5",
  prompt: systemPrompt,
  tools: [getUserLocation, getWeather],
  responseFormat,
  checkpointer,
});

// `thread_id` is a unique identifier for a given conversation.
const config = {
  configurable: { thread_id: "1" },
  context: { user_id: "1" },
};

const response = await agent.invoke(
  { messages: [{ role: "user", content: "what is the weather outside?" }] },
  config
);
console.log(response.structuredResponse);
// {
//   punny_response: "Florida is still having a 'sun-derful' day ...",
//   weather_conditions: "It's always sunny in Florida!"
// }

// Note that we can continue the conversation using the same `thread_id`.
const thankYouResponse = await agent.invoke(
  { messages: [{ role: "user", content: "thank you!" }] },
  config
);
console.log(thankYouResponse.structuredResponse);
// {
//   punny_response: "You're 'thund-erfully' welcome! ...",
//   weather_conditions: undefined
// }
恭喜！您现在拥有了一个能够：
  • 理解上下文并记住对话
  • 智能使用多个工具
  • 以一致格式提供结构化响应
  • 通过上下文处理特定用户信息
  • 在交互中维持对话状态
在GitHub上编辑此页面的源代码。
通过MCP将这些文档编程连接到Claude、VSCode等，以获取实时答案。