工作流与智能体

本指南回顾了常见的工作流和智能体模式。

• 工作流有预定的代码路径，被设计为按特定顺序运行。
• 智能体是动态的，定义自己的流程和工具使用。

LangGraph offers several benefits when building agents and workflows, including persistence, streaming, and support for debugging as well as deployment.

Trace and compare these workflow patterns with LangSmith. Follow the tracing quickstart to see how data flows through each step.

设置

To build a workflow or agent, you can use any chat model that supports structured outputs and tool calling. The following example uses Anthropic:

1. Install dependencies

npm

npm install @langchain/langgraph @langchain/core

2. Initialize the LLM:

import { ChatAnthropic } from "@langchain/anthropic";

const llm = new ChatAnthropic({
  model: "claude-sonnet-4-6",
  apiKey: "<your_anthropic_key>"
});

大语言模型(LLM)和增强

Workflows and agentic systems are based on LLMs and the various augmentations you add to them. Tool calling, structured outputs, and short term memory are a few options for tailoring LLMs to your needs.

import * as z from "zod";
import { tool } from "langchain";

// Schema for structured output
const SearchQuery = z.object({
  search_query: z.string().describe("Query that is optimized web search."),
  justification: z
    .string()
    .describe("Why this query is relevant to the user's request."),
});

// Augment the LLM with schema for structured output
const structuredLlm = llm.withStructuredOutput(SearchQuery);

// Invoke the augmented LLM
const output = await structuredLlm.invoke(
  "How does Calcium CT score relate to high cholesterol?"
);

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

// Augment the LLM with tools
const llmWithTools = llm.bindTools([multiply]);

// Invoke the LLM with input that triggers the tool call
const msg = await llmWithTools.invoke("What is 2 times 3?");

// Get the tool call
console.log(msg.tool_calls);

提示链

Prompt chaining is when each LLM call processes the output of the previous call. It’s often used for performing well-defined tasks that can be broken down into smaller, verifiable steps. Some examples include:

• Translating documents into different languages
• Verifying generated content for consistency

Graph API

import { StateGraph, StateSchema, GraphNode, ConditionalEdgeRouter } from "@langchain/langgraph";
import { z } from "zod/v4";

// Graph state
const State = new StateSchema({
  topic: z.string(),
  joke: z.string(),
  improvedJoke: z.string(),
  finalJoke: z.string(),
});

// Define node functions

// First LLM call to generate initial joke
const generateJoke: GraphNode<typeof State> = async (state) => {
  const msg = await llm.invoke(`Write a short joke about ${state.topic}`);
  return { joke: msg.content };
};

// Gate function to check if the joke has a punchline
const checkPunchline: ConditionalEdgeRouter<typeof State, "improveJoke"> = (state) => {
  // Simple check - does the joke contain "?" or "!"
  if (state.joke?.includes("?") || state.joke?.includes("!")) {
    return "Pass";
  }
  return "Fail";
};

// Second LLM call to improve the joke
const improveJoke: GraphNode<typeof State> = async (state) => {
  const msg = await llm.invoke(
    `Make this joke funnier by adding wordplay: ${state.joke}`
  );
  return { improvedJoke: msg.content };
};

// Third LLM call for final polish
const polishJoke: GraphNode<typeof State> = async (state) => {
  const msg = await llm.invoke(
    `Add a surprising twist to this joke: ${state.improvedJoke}`
  );
  return { finalJoke: msg.content };
};

// Build workflow
const chain = new StateGraph(State)
  .addNode("generateJoke", generateJoke)
  .addNode("improveJoke", improveJoke)
  .addNode("polishJoke", polishJoke)
  .addEdge("__start__", "generateJoke")
  .addConditionalEdges("generateJoke", checkPunchline, {
    Pass: "improveJoke",
    Fail: "__end__"
  })
  .addEdge("improveJoke", "polishJoke")
  .addEdge("polishJoke", "__end__")
  .compile();

// Invoke
const state = await chain.invoke({ topic: "cats" });
console.log("Initial joke:");
console.log(state.joke);
console.log("\n--- --- ---\n");
if (state.improvedJoke !== undefined) {
  console.log("Improved joke:");
  console.log(state.improvedJoke);
  console.log("\n--- --- ---\n");

  console.log("Final joke:");
  console.log(state.finalJoke);
} else {
  console.log("Joke failed quality gate - no punchline detected!");
}

并行化

With parallelization, LLMs work simultaneously on a task. This is either done by running multiple independent subtasks at the same time, or running the same task multiple times to check for different outputs. Parallelization is commonly used to:

• Split up subtasks and run them in parallel, which increases speed
• Run tasks multiple times to check for different outputs, which increases confidence

Some examples include:

• Running one subtask that processes a document for keywords, and a second subtask to check for formatting errors
• Running a task multiple times that scores a document for accuracy based on different criteria, like the number of citations, the number of sources used, and the quality of the sources

Graph API

import { StateGraph, StateSchema, GraphNode } from "@langchain/langgraph";
import * as z from "zod";

// Graph state
const State = new StateSchema({
  topic: z.string(),
  joke: z.string(),
  story: z.string(),
  poem: z.string(),
  combinedOutput: z.string(),
});

// Nodes
// First LLM call to generate initial joke
const callLlm1: GraphNode<typeof State> = async (state) => {
  const msg = await llm.invoke(`Write a joke about ${state.topic}`);
  return { joke: msg.content };
};

// Second LLM call to generate story
const callLlm2: GraphNode<typeof State> = async (state) => {
  const msg = await llm.invoke(`Write a story about ${state.topic}`);
  return { story: msg.content };
};

// Third LLM call to generate poem
const callLlm3: GraphNode<typeof State> = async (state) => {
  const msg = await llm.invoke(`Write a poem about ${state.topic}`);
  return { poem: msg.content };
};

// Combine the joke, story and poem into a single output
const aggregator: GraphNode<typeof State> = async (state) => {
  const combined = `Here's a story, joke, and poem about ${state.topic}!\n\n` +
    `STORY:\n${state.story}\n\n` +
    `JOKE:\n${state.joke}\n\n` +
    `POEM:\n${state.poem}`;
  return { combinedOutput: combined };
};

// Build workflow
const parallelWorkflow = new StateGraph(State)
  .addNode("callLlm1", callLlm1)
  .addNode("callLlm2", callLlm2)
  .addNode("callLlm3", callLlm3)
  .addNode("aggregator", aggregator)
  .addEdge("__start__", "callLlm1")
  .addEdge("__start__", "callLlm2")
  .addEdge("__start__", "callLlm3")
  .addEdge("callLlm1", "aggregator")
  .addEdge("callLlm2", "aggregator")
  .addEdge("callLlm3", "aggregator")
  .addEdge("aggregator", "__end__")
  .compile();

// Invoke
const result = await parallelWorkflow.invoke({ topic: "cats" });
console.log(result.combinedOutput);

路由

Routing workflows process inputs and then directs them to context-specific tasks. This allows you to define specialized flows for complex tasks. For example, a workflow built to answer product related questions might process the type of question first, and then route the request to specific processes for pricing, refunds, returns, etc.

Graph API

import { StateGraph, StateSchema, GraphNode, ConditionalEdgeRouter } from "@langchain/langgraph";
import * as z from "zod";

// Schema for structured output to use as routing logic
const routeSchema = z.object({
  step: z.enum(["poem", "story", "joke"]).describe(
    "The next step in the routing process"
  ),
});

// Augment the LLM with schema for structured output
const router = llm.withStructuredOutput(routeSchema);

// Graph state
const State = new StateSchema({
  input: z.string(),
  decision: z.string(),
  output: z.string(),
});

// Nodes
// Write a story
const llmCall1: GraphNode<typeof State> = async (state) => {
  const result = await llm.invoke([{
    role: "system",
    content: "You are an expert storyteller.",
  }, {
    role: "user",
    content: state.input
  }]);
  return { output: result.content };
};

// Write a joke
const llmCall2: GraphNode<typeof State> = async (state) => {
  const result = await llm.invoke([{
    role: "system",
    content: "You are an expert comedian.",
  }, {
    role: "user",
    content: state.input
  }]);
  return { output: result.content };
};

// Write a poem
const llmCall3: GraphNode<typeof State> = async (state) => {
  const result = await llm.invoke([{
    role: "system",
    content: "You are an expert poet.",
  }, {
    role: "user",
    content: state.input
  }]);
  return { output: result.content };
};

const llmCallRouter: GraphNode<typeof State> = async (state) => {
  // Route the input to the appropriate node
  const decision = await router.invoke([
    {
      role: "system",
      content: "Route the input to story, joke, or poem based on the user's request."
    },
    {
      role: "user",
      content: state.input
    },
  ]);

  return { decision: decision.step };
};

// Conditional edge function to route to the appropriate node
const routeDecision: ConditionalEdgeRouter<typeof State, "llmCall1" | "llmCall2" | "llmCall3"> = (state) => {
  // Return the node name you want to visit next
  if (state.decision === "story") {
    return "llmCall1";
  } else if (state.decision === "joke") {
    return "llmCall2";
  } else {
    return "llmCall3";
  }
};

// Build workflow
const routerWorkflow = new StateGraph(State)
  .addNode("llmCall1", llmCall1)
  .addNode("llmCall2", llmCall2)
  .addNode("llmCall3", llmCall3)
  .addNode("llmCallRouter", llmCallRouter)
  .addEdge("__start__", "llmCallRouter")
  .addConditionalEdges(
    "llmCallRouter",
    routeDecision,
    ["llmCall1", "llmCall2", "llmCall3"],
  )
  .addEdge("llmCall1", "__end__")
  .addEdge("llmCall2", "__end__")
  .addEdge("llmCall3", "__end__")
  .compile();

// Invoke
const state = await routerWorkflow.invoke({
  input: "Write me a joke about cats"
});
console.log(state.output);

编排器-工作器

In an orchestrator-worker configuration, the orchestrator:

• Breaks down tasks into subtasks
• Delegates subtasks to workers
• Synthesizes worker outputs into a final result

Orchestrator-worker workflows provide more flexibility and are often used when subtasks cannot be predefined the way they can with parallelization. This is common with workflows that write code or need to update content across multiple files. For example, a workflow that needs to update installation instructions for multiple Python libraries across an unknown number of documents might use this pattern.

Graph API

type SectionSchema = {
    name: string;
    description: string;
}
type SectionsSchema = {
    sections: SectionSchema[];
}

// Augment the LLM with schema for structured output
const planner = llm.withStructuredOutput(sectionsSchema);

Creating workers in LangGraph

Orchestrator-worker workflows are common and LangGraph has built-in support for them. The Send API lets you dynamically create worker nodes and send them specific inputs. Each worker has its own state, and all worker outputs are written to a shared state key that is accessible to the orchestrator graph. This gives the orchestrator access to all worker output and allows it to synthesize them into a final output. The example below iterates over a list of sections and uses the Send API to send a section to each worker.

import { StateGraph, StateSchema, ReducedValue, GraphNode, Send } from "@langchain/langgraph";
import * as z from "zod";

// Graph state
const State = new StateSchema({
  topic: z.string(),
  sections: z.array(z.custom<SectionsSchema>()),
  completedSections: new ReducedValue(
    z.array(z.string()).default(() => []),
    { reducer: (a, b) => a.concat(b) }
  ),
  finalReport: z.string(),
});

// Worker state
const WorkerState = new StateSchema({
  section: z.custom<SectionsSchema>(),
  completedSections: new ReducedValue(
    z.array(z.string()).default(() => []),
    { reducer: (a, b) => a.concat(b) }
  ),
});

// Nodes
const orchestrator: GraphNode<typeof State> = async (state) => {
  // Generate queries
  const reportSections = await planner.invoke([
    { role: "system", content: "Generate a plan for the report." },
    { role: "user", content: `Here is the report topic: ${state.topic}` },
  ]);

  return { sections: reportSections.sections };
};

const llmCall: GraphNode<typeof WorkerState> = async (state) => {
  // Generate section
  const section = await llm.invoke([
    {
      role: "system",
      content: "Write a report section following the provided name and description. Include no preamble for each section. Use markdown formatting.",
    },
    {
      role: "user",
      content: `Here is the section name: ${state.section.name} and description: ${state.section.description}`,
    },
  ]);

  // Write the updated section to completed sections
  return { completedSections: [section.content] };
};

const synthesizer: GraphNode<typeof State> = async (state) => {
  // List of completed sections
  const completedSections = state.completedSections;

  // Format completed section to str to use as context for final sections
  const completedReportSections = completedSections.join("\n\n---\n\n");

  return { finalReport: completedReportSections };
};

// Conditional edge function to create llm_call workers that each write a section of the report
const assignWorkers: ConditionalEdgeRouter<typeof State, "llmCall"> = (state) => {
  // Kick off section writing in parallel via Send() API
  return state.sections.map((section) =>
    new Send("llmCall", { section })
  );
};

// Build workflow
const orchestratorWorker = new StateGraph(State)
  .addNode("orchestrator", orchestrator)
  .addNode("llmCall", llmCall)
  .addNode("synthesizer", synthesizer)
  .addEdge("__start__", "orchestrator")
  .addConditionalEdges(
    "orchestrator",
    assignWorkers,
    ["llmCall"]
  )
  .addEdge("llmCall", "synthesizer")
  .addEdge("synthesizer", "__end__")
  .compile();

// Invoke
const state = await orchestratorWorker.invoke({
  topic: "Create a report on LLM scaling laws"
});
console.log(state.finalReport);

评估器-优化器

In evaluator-optimizer workflows, one LLM call creates a response and the other evaluates that response. If the evaluator or a human-in-the-loop determines the response needs refinement, feedback is provided and the response is recreated. This loop continues until an acceptable response is generated.

Evaluator-optimizer workflows are commonly used when there’s particular success criteria for a task, but iteration is required to meet that criteria. For example, there’s not always a perfect match when translating text between two languages. It might take a few iterations to generate a translation with the same meaning across the two languages.

Graph API

import { StateGraph, StateSchema, GraphNode, ConditionalEdgeRouter } from "@langchain/langgraph";
import * as z from "zod";

// Graph state
const State = new StateSchema({
  joke: z.string(),
  topic: z.string(),
  feedback: z.string(),
  funnyOrNot: z.string(),
});

// Schema for structured output to use in evaluation
const feedbackSchema = z.object({
  grade: z.enum(["funny", "not funny"]).describe(
    "Decide if the joke is funny or not."
  ),
  feedback: z.string().describe(
    "If the joke is not funny, provide feedback on how to improve it."
  ),
});

// Augment the LLM with schema for structured output
const evaluator = llm.withStructuredOutput(feedbackSchema);

// Nodes
const llmCallGenerator: GraphNode<typeof State> = async (state) => {
  // LLM generates a joke
  let msg;
  if (state.feedback) {
    msg = await llm.invoke(
      `Write a joke about ${state.topic} but take into account the feedback: ${state.feedback}`
    );
  } else {
    msg = await llm.invoke(`Write a joke about ${state.topic}`);
  }
  return { joke: msg.content };
};

const llmCallEvaluator: GraphNode<typeof State> = async (state) => {
  // LLM evaluates the joke
  const grade = await evaluator.invoke(`Grade the joke ${state.joke}`);
  return { funnyOrNot: grade.grade, feedback: grade.feedback };
};

// Conditional edge function to route back to joke generator or end based upon feedback from the evaluator
const routeJoke: ConditionalEdgeRouter<typeof State, "llmCallGenerator"> = (state) => {
  // Route back to joke generator or end based upon feedback from the evaluator
  if (state.funnyOrNot === "funny") {
    return "Accepted";
  } else {
    return "Rejected + Feedback";
  }
};

// Build workflow
const optimizerWorkflow = new StateGraph(State)
  .addNode("llmCallGenerator", llmCallGenerator)
  .addNode("llmCallEvaluator", llmCallEvaluator)
  .addEdge("__start__", "llmCallGenerator")
  .addEdge("llmCallGenerator", "llmCallEvaluator")
  .addConditionalEdges(
    "llmCallEvaluator",
    routeJoke,
    {
      // Name returned by routeJoke : Name of next node to visit
      "Accepted": "__end__",
      "Rejected + Feedback": "llmCallGenerator",
    }
  )
  .compile();

// Invoke
const state = await optimizerWorkflow.invoke({ topic: "Cats" });
console.log(state.joke);

智能体

Agents are typically implemented as an LLM performing actions using tools. They operate in continuous feedback loops, and are used in situations where problems and solutions are unpredictable. Agents have more autonomy than workflows, and can make decisions about the tools they use and how to solve problems. You can still define the available toolset and guidelines for how agents behave.

To get started with agents, see the quickstart or read more about how they work in LangChain.

Using tools

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

// Define tools
const multiply = tool(
  ({ a, b }) => {
    return a * b;
  },
  {
    name: "multiply",
    description: "Multiply two numbers together",
    schema: z.object({
      a: z.number().describe("first number"),
      b: z.number().describe("second number"),
    }),
  }
);

const add = tool(
  ({ a, b }) => {
    return a + b;
  },
  {
    name: "add",
    description: "Add two numbers together",
    schema: z.object({
      a: z.number().describe("first number"),
      b: z.number().describe("second number"),
    }),
  }
);

const divide = tool(
  ({ a, b }) => {
    return 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 tools = [add, multiply, divide];
const toolsByName = Object.fromEntries(tools.map((tool) => [tool.name, tool]));
const llmWithTools = llm.bindTools(tools);

Graph API

import { StateGraph, StateSchema, MessagesValue, GraphNode, ConditionalEdgeRouter } from "@langchain/langgraph";
import { ToolNode } from "@langchain/langgraph/prebuilt";
import {
  SystemMessage,
  ToolMessage
} from "@langchain/core/messages";

// Graph state
const State = new StateSchema({
  messages: MessagesValue,
});

// Nodes
const llmCall: GraphNode<typeof State> = async (state) => {
  // LLM decides whether to call a tool or not
  const result = await llmWithTools.invoke([
    {
      role: "system",
      content: "You are a helpful assistant tasked with performing arithmetic on a set of inputs."
    },
    ...state.messages
  ]);

  return {
    messages: [result]
  };
};

const toolNode = new ToolNode(tools);

// Conditional edge function to route to the tool node or end
const shouldContinue: ConditionalEdgeRouter<typeof State, "toolNode"> = (state) => {
  const messages = state.messages;
  const lastMessage = messages.at(-1);

  // 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__";
};

// Build workflow
const agentBuilder = new StateGraph(State)
  .addNode("llmCall", llmCall)
  .addNode("toolNode", toolNode)
  // Add edges to connect nodes
  .addEdge("__start__", "llmCall")
  .addConditionalEdges(
    "llmCall",
    shouldContinue,
    ["toolNode", "__end__"]
  )
  .addEdge("toolNode", "llmCall")
  .compile();

// Invoke
const messages = [{
  role: "user",
  content: "Add 3 and 4."
}];
const result = await agentBuilder.invoke({ messages });
console.log(result.messages);

ToolNode

ToolNode is a prebuilt node that executes tools in LangGraph workflows. It handles parallel tool execution, error handling, and state injection automatically.

Use ToolNode when you need fine-grained control over how your graph executes tools. This is the building block that powers tool execution in many LangGraph agent patterns.

import { ToolNode } from "@langchain/langgraph/prebuilt";
import { tool } from "@langchain/core/tools";
import * as z from "zod";

const search = tool(
  ({ query }) => `Results for: ${query}`,
  {
    name: "search",
    description: "Search for information.",
    schema: z.object({ query: z.string() }),
  }
);

const calculator = tool(
  ({ expression }) => String(eval(expression)),
  {
    name: "calculator",
    description: "Evaluate a math expression.",
    schema: z.object({ expression: z.string() }),
  }
);

const toolNode = new ToolNode([search, calculator]);

---

将这些文档连接到 Claude、VSCode 等，通过 MCP 获取实时答案。

在 GitHub 上编辑此页面或提交问题。