Introduction

• playground  you can see the demo code on the playground.

Set up a fully functional Crawler Agent powered by OpenAI’s GPT model quickly using the Agentica CLI.

Quick CLI Setup

Launch the Agentica Setup Wizard with a single command:

npx agentica start crawler-agent

The wizard guides you through:

• Installing required packages (e.g., [email protected])
• Choosing your package manager and project type
• Selecting the WEB_CRAWLER controller
• Entering your OPENAI_API_KEY

Once complete, Agentica automatically generates your code, creates a .env file, and installs all dependencies.

Generated Code Overview

The generated code looks like this:

import { Agentica } from "@agentica/core";
import typia from "typia";
import dotenv from "dotenv";
import OpenAI from "openai";
 
import { WebCrawlerService } from "@wrtnlabs/connector-web-crawler";
 
dotenv.config();
 
const agent = new Agentica({
  model: "chatgpt",
  vendor: {
    api: new OpenAI({ apiKey: process.env.OPENAI_API_KEY! }),
    model: "gpt-4o-mini",
  },
  controllers: [
    {
      name: "Web Crawler Connector",
      protocol: "class",
      application: typia.llm.application<WebCrawlerService, "chatgpt">(),
      execute: new WebCrawlerService({
        zenRowsApiKey: process.env.ZENROWS_API_KEY!,
      }),
    },
  ],
});
 
const main = async () => {
  console.log(await agent.conversate("What can you do?"));
};
 
main();

This code instantly sets up your Web Crawler Agent, ready to intelligently crawl the web.

What This Does

Your Web Crawler Agent will:

• Process Arxiv Search Data: Interact with the Arxiv Search API using the WebCrawlerService.
• Leverage OpenAI’s GPT Model: Use OpenAI’s GPT model to generate responses to user queries.
• Automate Web Crawling: Automatically crawl the web to gather data for your project.
• Securely Store API Keys: Store your API keys securely in a .env file.

Set up your environment variables in the .env file:

OPENAI_API_KEY=your-openai-api-key
ZENROWS_API_KEY=your-zenrows-api-key

Use Case Example

Imagine asking the agent:

“Crawl this https://wrtnlabs.io/agentica/docs/  and summarize the content.”

Available Functions

For a complete list of available functions in WebCrawlerService, check out the source code: 👉 wrtnlabs/connectors - WebCrawlerService.ts 

Your AI-powered Web Crawler Agent is ready to crawl the web and gather data for your project. 🚀