Web Scraping

Extract structured data from websites

Overview

Web scraping lets you extract structured data from websites. Browserbase provides a reliable browser infrastructure that helps you build scrapers that can:

• Scale without infrastructure management
• Maintain consistent performance
• Avoid bot detection and CAPTCHAs with Browserbase’s stealth mode
• Provide debugging and monitoring tools with session replays and live views

This guide will help you get started with web scraping on Browserbase and highlight best practices.

Scraping a website

Using a sample website, we’ll scrape the title, price, and some other details of books from the website.

Follow Along: Web Scraping Example

Step-by-step code for web scraping

Code Example

Node.js

Stagehand

import { Stagehand } from "@browserbasehq/stagehand";
import { z } from "zod";
import dotenv from "dotenv";
dotenv.config();

const stagehand = new Stagehand({
    env: "BROWSERBASE",
    verbose: 0,
});

async function scrapeBooks() {
    await stagehand.init();
    const page = stagehand.page;

    await page.goto("https://books.toscrape.com/");

    const scrape = await page.extract({
        instruction: "Extract the books from the page",
        schema: z.object({
            books: z.array(z.object({
                title: z.string(),
                price: z.string(),
                image: z.string(),
                inStock: z.string(),
                link: z.string(),
            }))
        }),
    });

    console.log(scrape.books);

    await stagehand.close();
    return books;
}

const books = scrapeBooks().catch(console.error);

Stagehand

import { Stagehand } from "@browserbasehq/stagehand";
import { z } from "zod";
import dotenv from "dotenv";
dotenv.config();

const stagehand = new Stagehand({
    env: "BROWSERBASE",
    verbose: 0,
});

async function scrapeBooks() {
    await stagehand.init();
    const page = stagehand.page;

    await page.goto("https://books.toscrape.com/");

    const scrape = await page.extract({
        instruction: "Extract the books from the page",
        schema: z.object({
            books: z.array(z.object({
                title: z.string(),
                price: z.string(),
                image: z.string(),
                inStock: z.string(),
                link: z.string(),
            }))
        }),
    });

    console.log(scrape.books);

    await stagehand.close();
    return books;
}

const books = scrapeBooks().catch(console.error);

Playwright

import os
from playwright.sync_api import sync_playwright
from browserbase import Browserbase
from dotenv import load_dotenv

load_dotenv()

def create_session():
    """Creates a Browserbase session."""
    bb = Browserbase(api_key=os.environ["BROWSERBASE_API_KEY"])
    session = bb.sessions.create(
        project_id=os.environ["BROWSERBASE_PROJECT_ID"],
        # Add configuration options here if needed
    )
    return session

def web_scrape():
    """Automates form filling using Playwright with Browserbase."""
    session = create_session()
    print(f"View session replay at https://browserbase.com/sessions/{session.id}")

    with sync_playwright() as p:
        browser = p.chromium.connect_over_cdp(session.connect_url)

        # Get the default browser context and page
        context = browser.contexts[0]
        page = context.pages[0]

        # Navigate to the form page
        page.goto("https://books.toscrape.com/")

        # Extract the books from the page
        items = page.locator('article.product_pod')
        books = items.all()

        book_data_list = []
        for book in books:

            book_data = {
                "title": book.locator('h3 a').get_attribute('title'),
                "price": book.locator('p.price_color').text_content(),
                "image": book.locator('div.image_container img').get_attribute('src'),
                "inStock": book.locator('p.instock.availability').text_content().strip(),
                "link": book.locator('h3 a').get_attribute('href')
            }
            
            book_data_list.append(book_data)

        print("Shutting down...")
        page.close()
        browser.close()

        return book_data_list

if __name__ == "__main__":
    books = web_scrape()
    print(books)

Example output

[
  {
    title: 'A Light in the Attic',
    price: '£51.77',
    image: 'https://books.toscrape.com/media/cache/2c/da/2cdad67c44b002e7ead0cc35693c0e8b.jpg',
    inStock: 'In stock',
    link: 'catalogue/a-light-in-the-attic_1000/index.html'
  },
  ...
]

Best Practices for Web Scraping

Follow these best practices to build reliable, efficient, and ethical web scrapers with Browserbase.

Ethical Scraping

• Respect robots.txt: Check the website’s robots.txt file for crawling guidelines
• Rate limiting: Implement reasonable delays between requests (2-5 seconds)
• Terms of Service: Review the website’s terms of service before scraping
• Data usage: Only collect and use data in accordance with the website’s policies

Performance Optimization

• Batch processing: Process multiple pages in batches with concurrent sessions
• Selective scraping: Only extract the data you need
• Resource management: Close browser sessions promptly after use
• Connection reuse: Reuse browsers for sequential scraping tasks

Stealth and Anti-Bot Avoidance

• Enable Browserbase Advanced Stealth mode: Helps avoid bot detection
• Randomize behavior: Add variable delays between actions
• Use proxies: Rotate IPs to distribute requests
• Mimic human interaction: Add realistic mouse movements and delays
• Handle CAPTCHAs: Enable Browserbase’s automatic CAPTCHA solving

Next Steps

Now that you understand the basics of web scraping with Browserbase, here are some features to explore next:

Stealth Mode

Configure fingerprinting and CAPTCHA solving

Browser Contexts

Persist cookies and session data

Proxies

Configure IP rotation and geolocation

Assistant

Responses are generated using AI and may contain mistakes.