LangChain Web Research Agent: Complete Tutorial

LangChain makes it easy to build AI agents with custom tools. This tutorial shows you how to create a web research agent that can search, read, and synthesize information from the web.

Architecture

User Query → Agent (GPT-4) → Tools (Search, Read) → Synthesis → Response

Step 1: Install Dependencies

npm install langchain @langchain/openai @langchain/core

Step 2: Create Tryb Tools

import { DynamicTool } from "langchain/tools";

const searchTool = new DynamicTool({
  name: "web_search",
  description: "Search the web for information. Input should be a search query.",
  func: async (query: string) => {
    const res = await fetch("https://api.tryb.dev/v1/search", {
      method: "POST",
      headers: {
        "Authorization": `Bearer ${process.env.TRYB_API_KEY}`,
        "Content-Type": "application/json"
      },
      body: JSON.stringify({ query, num_results: 5 })
    });
    const data = await res.json();
    return JSON.stringify(data.results);
  }
});

const readTool = new DynamicTool({
  name: "read_webpage",
  description: "Read the content of a webpage. Input should be a URL.",
  func: async (url: string) => {
    const res = await fetch("https://api.tryb.dev/v1/read", {
      method: "POST",
      headers: {
        "Authorization": `Bearer ${process.env.TRYB_API_KEY}`,
        "Content-Type": "application/json"
      },
      body: JSON.stringify({ url })
    });
    const data = await res.json();
    return data.data.markdown;
  }
});

Step 3: Create the Agent

import { ChatOpenAI } from "@langchain/openai";
import { AgentExecutor, createOpenAIFunctionsAgent } from "langchain/agents";
import { ChatPromptTemplate } from "@langchain/core/prompts";

const llm = new ChatOpenAI({ model: "gpt-4-turbo-preview" });

const prompt = ChatPromptTemplate.fromMessages([
  ["system", `You are a research assistant. Use the web_search tool to find relevant pages, then use read_webpage to get detailed content. Synthesize findings into a comprehensive answer.`],
  ["human", "{input}"],
  ["placeholder", "{agent_scratchpad}"]
]);

const agent = await createOpenAIFunctionsAgent({
  llm,
  tools: [searchTool, readTool],
  prompt
});

const executor = new AgentExecutor({ agent, tools: [searchTool, readTool] });

Step 4: Run Research Queries

const result = await executor.invoke({
  input: "What are the latest developments in quantum computing?"
});

console.log(result.output);

Advanced: Adding Memory

import { BufferMemory } from "langchain/memory";

const memory = new BufferMemory({
  memoryKey: "chat_history",
  returnMessages: true
});

const executorWithMemory = new AgentExecutor({
  agent,
  tools: [searchTool, readTool],
  memory
});

Best Practices

• Limit tool calls to prevent runaway costs
• Cache frequently accessed URLs
• Add timeout handling for long-running research
• Log all tool calls for debugging

Full Source Code

Get the complete implementation: GitHub Repository