​
概述

在本教程中，我们将使用 LangGraph 构建一个检索智能体。 LangChain 提供内置的智能体实现，使用 LangGraph 原语实现。如果需要更深入的自定义，可以直接在 LangGraph 中实现智能体。本指南演示了检索智能体的示例实现。当您希望 LLM 决定是从向量存储中检索上下文还是直接响应用户时，检索智能体非常有用。 在本教程结束时，我们将完成以下工作：
  1. 获取和预处理将用于检索的文档。
  2. 为这些文档建立索引以进行语义搜索，并为智能体创建检索器工具。
  3. 构建一个可以决定何时使用检索器工具的智能体 RAG 系统。
混合 RAG

​
概念

我们将涵盖以下概念：

​
设置

让我们下载所需的包并设置我们的 API 密钥： 
npm install @langchain/langgraph @langchain/openai @langchain/community @langchain/textsplitters
注册 LangSmith 以快速发现问题并提高 LangGraph 项目的性能。LangSmith 让您可以使用跟踪数据来调试、测试和监控使用 LangGraph 构建的 LLM 应用程序。

​
1. 预处理文档

  1. 获取要在我们的 RAG 系统中使用的文档。我们将使用 Lilian Weng 优秀博客中最近的三篇文章。我们将首先使用 CheerioWebBaseLoader 获取页面内容：
import { CheerioWebBaseLoader } from "@langchain/community/document_loaders/web/cheerio";

const urls = [
  "https://lilianweng.github.io/posts/2023-06-23-agent/",
  "https://lilianweng.github.io/posts/2023-03-15-prompt-engineering/",
  "https://lilianweng.github.io/posts/2023-10-25-adv-attack-llm/",
];

const docs = await Promise.all(
  urls.map((url) => new CheerioWebBaseLoader(url).load()),
);
  1. 将获取的文档拆分为更小的块以索引到我们的向量存储中：
import { RecursiveCharacterTextSplitter } from "@langchain/textsplitters";

const docsList = docs.flat();

const textSplitter = new RecursiveCharacterTextSplitter({
  chunkSize: 500,
  chunkOverlap: 50,
});
const docSplits = await textSplitter.splitDocuments(docsList);

​
2. 创建检索器工具

现在我们有了拆分的文档，我们可以将它们索引到向量存储中，用于语义搜索。
  1. 使用内存向量存储和 OpenAI 嵌入：
import { MemoryVectorStore } from "@langchain/classic/vectorstores/memory";
import { OpenAIEmbeddings } from "@langchain/openai";

const vectorStore = await MemoryVectorStore.fromDocuments(
  docSplits,
  new OpenAIEmbeddings(),
);

const retriever = vectorStore.asRetriever();
  1. 使用 LangChain 的预构建 createRetrieverTool 创建检索器工具：
import { createRetrieverTool } from "@langchain/classic/tools/retriever";

const tool = createRetrieverTool(
  retriever,
  {
    name: "retrieve_blog_posts",
    description:
      "Search and return information about Lilian Weng blog posts on LLM agents, prompt engineering, and adversarial attacks on LLMs.",
  },
);
const tools = [tool];

​
3. 生成查询

现在我们将开始为我们的智能 RAG 图构建组件（节点和边）。
  1. 构建 generateQueryOrRespond 节点。它将调用 LLM 根据当前图状态（消息列表）生成响应。给定输入消息，它将决定使用检索器工具进行检索，或直接响应用户。请注意，我们通过 .bindTools 让聊天模型访问我们之前创建的 tools：
import { ChatOpenAI } from "@langchain/openai";

async function generateQueryOrRespond(state) {
  const { messages } = state;
  const model = new ChatOpenAI({
    model: "gpt-4o",
    temperature: 0,
  }).bindTools(tools);  

  const response = await model.invoke(messages);
  return {
    messages: [response],
  };
}
  1. 在随机输入上尝试它：
import { HumanMessage } from "@langchain/core/messages";

const input = { messages: [new HumanMessage("hello!")] };
const result = await generateQueryOrRespond(input);
console.log(result.messages[0]);
Output: 
AIMessage {
  content: "Hello! How can I help you today?",
  tool_calls: []
}
  1. 提出需要语义搜索的问题：
const input = {
  messages: [
    new HumanMessage("What does Lilian Weng say about types of reward hacking?")
  ]
};
const result = await generateQueryOrRespond(input);
console.log(result.messages[0]);
Output: 
AIMessage {
  content: "",
  tool_calls: [
    {
      name: "retrieve_blog_posts",
      args: { query: "types of reward hacking" },
      id: "call_...",
      type: "tool_call"
    }
  ]
}

​
4. 评分文档

  1. 添加节点 — gradeDocuments — 以确定检索的文档是否与问题相关。我们将使用 Zod 的结构化输出模型进行文档评分。我们还将添加条件边 — checkRelevance — 检查评分结果并返回要去的节点的名称（generate 或 rewrite）：
import * as z from "zod";
import { ChatPromptTemplate } from "@langchain/core/prompts";
import { ChatOpenAI } from "@langchain/openai";
import { AIMessage } from "@langchain/core/messages";

const prompt = ChatPromptTemplate.fromTemplate(
  `You are a grader assessing relevance of retrieved docs to a user question.
  Here are the retrieved docs:
  \n ------- \n
  {context}
  \n ------- \n
  Here is the user question: {question}
  If the content of the docs are relevant to the users question, score them as relevant.
  Give a binary score 'yes' or 'no' score to indicate whether the docs are relevant to the question.
  Yes: The docs are relevant to the question.
  No: The docs are not relevant to the question.`,
);

const gradeDocumentsSchema = z.object({
  binaryScore: z.string().describe("Relevance score 'yes' or 'no'"),  
})

async function gradeDocuments(state) {
  const { messages } = state;

  const model = new ChatOpenAI({
    model: "gpt-4o",
    temperature: 0,
  }).withStructuredOutput(gradeDocumentsSchema);

  const score = await prompt.pipe(model).invoke({
    question: messages.at(0)?.content,
    context: messages.at(-1)?.content,
  });

  if (score.binaryScore === "yes") {
    return "generate";
  }
  return "rewrite";
}
  1. 使用工具响应中的不相关文档运行此操作：
import { ToolMessage } from "@langchain/core/messages";

const input = {
  messages: [
      new HumanMessage("What does Lilian Weng say about types of reward hacking?"),
      new AIMessage({
          tool_calls: [
              {
                  type: "tool_call",
                  name: "retrieve_blog_posts",
                  args: { query: "types of reward hacking" },
                  id: "1",
              }
          ]
      }),
      new ToolMessage({
          content: "meow",
          tool_call_id: "1",
      })
  ]
}
const result = await gradeDocuments(input);
  1. 确认相关文档被正确分类：
const input = {
  messages: [
      new HumanMessage("What does Lilian Weng say about types of reward hacking?"),
      new AIMessage({
          tool_calls: [
              {
                  type: "tool_call",
                  name: "retrieve_blog_posts",
                  args: { query: "types of reward hacking" },
                  id: "1",
              }
          ]
      }),
      new ToolMessage({
          content: "reward hacking can be categorized into two types: environment or goal misspecification, and reward tampering",
          tool_call_id: "1",
      })
  ]
}
const result = await gradeDocuments(input);

​
5. Rewrite question

  1. 构建 rewrite 节点。检索器工具可能返回可能不相关的文档，这表明需要改进原始用户问题。为此，我们将调用 rewrite 节点：
import { ChatPromptTemplate } from "@langchain/core/prompts";
import { ChatOpenAI } from "@langchain/openai";

const rewritePrompt = ChatPromptTemplate.fromTemplate(
  `Look at the input and try to reason about the underlying semantic intent / meaning. \n
  Here is the initial question:
  \n ------- \n
  {question}
  \n ------- \n
  Formulate an improved question:`,
);

async function rewrite(state) {
  const { messages } = state;
  const question = messages.at(0)?.content;

  const model = new ChatOpenAI({
    model: "gpt-4o",
    temperature: 0,
  });

  const response = await rewritePrompt.pipe(model).invoke({ question });
  return {
    messages: [response],
  };
}
  1. 尝试一下：
import { HumanMessage, AIMessage, ToolMessage } from "@langchain/core/messages";

const input = {
  messages: [
    new HumanMessage("What does Lilian Weng say about types of reward hacking?"),
    new AIMessage({
      content: "",
      tool_calls: [
        {
          id: "1",
          name: "retrieve_blog_posts",
          args: { query: "types of reward hacking" },
          type: "tool_call"
        }
      ]
    }),
    new ToolMessage({ content: "meow", tool_call_id: "1" })
  ]
};

const response = await rewrite(input);
console.log(response.messages[0].content);
Output: 
What are the different types of reward hacking described by Lilian Weng, and how does she explain them?

​
6. Generate an answer

  1. 构建 generate 节点：如果我们通过评分器检查，我们可以根据原始问题和检索的上下文生成最终答案：
import { ChatPromptTemplate } from "@langchain/core/prompts";
import { ChatOpenAI } from "@langchain/openai";

async function generate(state) {
  const { messages } = state;
  const question = messages.at(0)?.content;
  const context = messages.at(-1)?.content;

  const prompt = ChatPromptTemplate.fromTemplate(
  `You are an assistant for question-answering tasks.
      Use the following pieces of retrieved context to answer the question.
      If you don't know the answer, just say that you don't know.
      Use three sentences maximum and keep the answer concise.
      Question: {question}
      Context: {context}`
  );

  const llm = new ChatOpenAI({
    model: "gpt-4o",
    temperature: 0,
  });

  const ragChain = prompt.pipe(llm);

  const response = await ragChain.invoke({
    context,
    question,
  });

  return {
    messages: [response],
  };
}
  1. 尝试它：
import { HumanMessage, AIMessage, ToolMessage } from "@langchain/core/messages";

const input = {
  messages: [
    new HumanMessage("What does Lilian Weng say about types of reward hacking?"),
    new AIMessage({
      content: "",
      tool_calls: [
        {
          id: "1",
          name: "retrieve_blog_posts",
          args: { query: "types of reward hacking" },
          type: "tool_call"
        }
      ]
    }),
    new ToolMessage({
      content: "reward hacking can be categorized into two types: environment or goal misspecification, and reward tampering",
      tool_call_id: "1"
    })
  ]
};

const response = await generate(input);
console.log(response.messages[0].content);
Output: 
Lilian Weng categorizes reward hacking into two types: environment or goal misspecification, and reward tampering. She considers reward hacking as a broad concept that includes both of these categories. Reward hacking occurs when an agent exploits flaws or ambiguities in the reward function to achieve high rewards without performing the intended behaviors.

​
7. Assemble the graph

现在我们将把所有节点和边组装成一个完整的图：
  • 从 generateQueryOrRespond 开始并确定是否需要调用检索器工具
  • 使用条件边路由到下一步：
    • 如果 generateQueryOrRespond 返回 tool_calls，调用检索器工具检索上下文
    • 否则，直接响应用户
  • 为问题的相关性对检索的文档内容进行评分（gradeDocuments）并路由到下一步：
    • 如果不相关，使用 rewrite 重写问题，然后再次调用 generateQueryOrRespond
    • 如果相关，继续到 generate 并使用带有检索的文档上下文的 @[ToolMessage] 生成最终响应
import { StateGraph, START, END } from "@langchain/langgraph";
import { ToolNode } from "@langchain/langgraph/prebuilt";
import { AIMessage } from "langchain";

// 为检索器创建 ToolNode
const toolNode = new ToolNode(tools);

// 辅助函数，确定是否应该检索
function shouldRetrieve(state) {
  const { messages } = state;
  const lastMessage = messages.at(-1);

  if (AIMessage.isInstance(lastMessage) && lastMessage.tool_calls.length) {
    return "retrieve";
  }
  return END;
}

// 定义图
const builder = new StateGraph(GraphState)
  .addNode("generateQueryOrRespond", generateQueryOrRespond)
  .addNode("retrieve", toolNode)
  .addNode("gradeDocuments", gradeDocuments)
  .addNode("rewrite", rewrite)
  .addNode("generate", generate)
  // 添加边
  .addEdge(START, "generateQueryOrRespond")
  // 决定是否检索
  .addConditionalEdges("generateQueryOrRespond", shouldRetrieve)
  .addEdge("retrieve", "gradeDocuments")
  // 评分文档后采取的边
  .addConditionalEdges(
    "gradeDocuments",
    // 根据评分决策路由
    (state) => {
      // gradeDocuments 函数返回 "generate" 或 "rewrite"
      const lastMessage = state.messages.at(-1);
      return lastMessage.content === "generate" ? "generate" : "rewrite";
    }
  )
  .addEdge("generate", END)
  .addEdge("rewrite", "generateQueryOrRespond");

// 编译
const graph = builder.compile();

​
8. Run the agentic RAG

现在让我们通过使用问题运行来测试完整的图： 
import { HumanMessage } from "@langchain/core/messages";

const inputs = {
  messages: [
    new HumanMessage("What does Lilian Weng say about types of reward hacking?")
  ]
};

for await (const output of await graph.stream(inputs)) {
  for (const [key, value] of Object.entries(output)) {
    const lastMsg = output[key].messages[output[key].messages.length - 1];
    console.log(`Output from node: '${key}'`);
    console.log({
      type: lastMsg._getType(),
      content: lastMsg.content,
      tool_calls: lastMsg.tool_calls,
    });
    console.log("---\n");
  }
}
Output: 
Output from node: 'generateQueryOrRespond'
{
  type: 'ai',
  content: '',
  tool_calls: [
    {
      name: 'retrieve_blog_posts',
      args: { query: 'types of reward hacking' },
      id: 'call_...',
      type: 'tool_call'
    }
  ]
}
---

Output from node: 'retrieve'
{
  type: 'tool',
  content: '(Note: Some work defines reward tampering as a distinct category...\n' +
    'At a high level, reward hacking can be categorized into two types: environment or goal misspecification, and reward tampering.\n' +
    '...',
  tool_calls: undefined
}
---

Output from node: 'generate'
{
  type: 'ai',
  content: 'Lilian Weng categorizes reward hacking into two types: environment or goal misspecification, and reward tampering. She considers reward hacking as a broad concept that includes both of these categories. Reward hacking occurs when an agent exploits flaws or ambiguities in the reward function to achieve high rewards without performing the intended behaviors.',
  tool_calls: []
}
---
Edit the source of this page on GitHub.
Connect these docs programmatically to Claude, VSCode, and more via MCP for real-time answers.