Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.langchain.com/llms.txt

Use this file to discover all available pages before exploring further.

This guide walks you through creating your first deep agent with planning, file system tools, and subagent capabilities. You’ll build a research agent that can conduct research and write reports.
Using an AI coding assistant?
  • Install the LangChain Docs MCP server to give your agent access to up-to-date LangChain documentation and examples.
  • Install LangChain Skills to improve your agent’s performance on LangChain ecosystem tasks.

​
Prerequisites

Before you begin, make sure you have an API key from a model provider (e.g., Gemini, Anthropic, OpenAI).
Deep Agents require a model that supports tool calling. See customization for how to configure your model.

​
Step 1: Install dependencies

npm install deepagents langchain @langchain/core @langchain/tavily
This guide uses Tavily as an example search provider, but you can substitute any search API (e.g., DuckDuckGo, SerpAPI, Brave Search).

​
Step 2: Set up your API keys

export GOOGLE_API_KEY="your-api-key"
export TAVILY_API_KEY="your-tavily-api-key"

​
Step 3: Create a search tool

import { tool } from "langchain";
import { TavilySearch } from "@langchain/tavily";
import { z } from "zod";

const internetSearch = tool(
  async ({
    query,
    maxResults = 5,
    topic = "general",
    includeRawContent = false,
  }: {
    query: string;
    maxResults?: number;
    topic?: "general" | "news" | "finance";
    includeRawContent?: boolean;
  }) => {
    const tavilySearch = new TavilySearch({
      maxResults,
      tavilyApiKey: process.env.TAVILY_API_KEY,
      includeRawContent,
      topic,
    });
    return await tavilySearch._call({ query });
  },
  {
    name: "internet_search",
    description: "Run a web search",
    schema: z.object({
      query: z.string().describe("The search query"),
      maxResults: z
        .number()
        .optional()
        .default(5)
        .describe("Maximum number of results to return"),
      topic: z
        .enum(["general", "news", "finance"])
        .optional()
        .default("general")
        .describe("Search topic category"),
      includeRawContent: z
        .boolean()
        .optional()
        .default(false)
        .describe("Whether to include raw content"),
    }),
  },
);

​
Step 4: Create a deep agent

import { createDeepAgent } from "deepagents";

// System prompt to steer the agent to be an expert researcher
const researchInstructions = `You are an expert researcher. Your job is to conduct thorough research and then write a polished report.

You have access to an internet search tool as your primary means of gathering information.

## \`internet_search\`

Use this to run an internet search for a given query. You can specify the max number of results to return, the topic, and whether raw content should be included.
`;
Pick a model from your provider. By default, createDeepAgent uses claude-sonnet-4-6. Pass a model string to use a different provider — see Suggested models for the full list. 
const agent = createDeepAgent({
  model: "google-genai:gemini-3.1-pro-preview",
  tools: [internetSearch],
  systemPrompt: researchInstructions,
});

​
Step 5: Run the agent

const result = await agent.invoke({
  messages: [{ role: "user", content: "What is langgraph?" }],
});

// Print the agent's response
console.log(result.messages[result.messages.length - 1].content);

​
How does it work?

Your deep agent automatically:
  1. Plans its approach using the built-in write_todos tool to break down the research task.
  2. Conducts research by calling the internet_search tool to gather information.
  3. Manages context by using file system tools (write_file, read_file) to offload large search results.
  4. Spawns subagents as needed to delegate complex subtasks to specialized subagents.
  5. Synthesizes a report to compile findings into a coherent response.

​
Examples

For agents, patterns, and applications you can build with Deep Agents, see Examples.

​
Streaming

Deep Agents have built-in streaming for real-time updates from agent execution using LangGraph. This allows you to observe output progressively and review and debug agent and subagent work, such as tool calls, tool results, and LLM responses.

​
Next steps

Now that you’ve built your first deep agent:
Connect these docs to Claude, VSCode, and more via MCP for real-time answers.
Edit this page on GitHub or file an issue.