BrowserBook

Introduction

Next →

Welcome to BrowserBook

BrowserBook is a developer-first IDE for building reliable web automations. If your workflows are locked inside legacy systems with no API or interoperability, BrowserBook unlocks the browser as an automation surface.

With BrowserBook you can:

  • Write automation logic in a Jupyter-style notebook.
  • Execute against a hosted browser with no local setup required.
  • Deploy notebooks to run via API or on a schedule [coming soon].
  • Integrate seamlessly with your agents and automation workflows.

We designed BrowserBook to solve the last mile problem for agents: when API coverage ends, BrowserBook lets your agent reliably interact with the web.

Quickstart

  1. 1
    Create a new notebook
    From the sidebar or notebooks list, click New Notebook.
  2. 2
    Write your first automation

    Each notebook is comprised of markdown text and TypeScript code arranged in one or more "cells". To add a new cell, click + Markdown or + Code at the top of the interface.

    In a code cell, you can write Playwright scripts to automate the inline browser instance. For example, to open a webpage and extract its title:

    await page.goto("https://example.com")
const title = await page.title()
console.log(title)
  3. 3
    Run the notebook
    Click Run All, or run a single cell with Shift+Enter or by clicking the Play button at the top of the cell. Your automation executes against an inline hosted browser session—no local setup needed.
  4. 4
    View results
    Outputs appear inline below each cell. You can also download artifacts (JSON, CSV, files) generated from the automation. [coming soon]
  5. 5