stagehandvsnodriver

MIT 183 14 22,012

2.8 million (month) Oct 29 2024 3.2.1(2026-04-10 21:10:37 ago)

4,003 2 14 AGPL-3.0

Jan 15 2024 321.9 thousand (month) 0.48.1(2025-11-09 05:57:23 ago)

Stagehand is an AI-powered browser automation framework for JavaScript and TypeScript, built by Browserbase. It provides a simple API for controlling browsers using natural language instructions, powered by large language models.

Stagehand offers three core primitives:

act() Performs actions on the page described in natural language. For example, page.act("click the login button") will find and click the appropriate element.
extract() Extracts structured data from the page based on a natural language description and an optional schema definition.
observe() Analyzes the current page state and returns actionable elements and their descriptions, useful for understanding what actions are available on a page.

Key features include:

TypeScript-first Built with full TypeScript support and type-safe extraction using Zod schemas.
Multiple LLM providers Works with OpenAI, Anthropic, and other LLM providers for powering the AI.
Vision and DOM analysis Combines visual screenshot analysis with DOM inspection for robust element identification.
Playwright integration Uses Playwright as the browser automation backend, giving access to the full Playwright API alongside AI-powered actions.
Browserbase cloud Optionally integrates with Browserbase cloud for managed browser infrastructure.

Stagehand is particularly suited for automating complex web workflows where traditional selectors would be fragile, such as interacting with frequently changing UIs or scraping sites with dynamic layouts.

nodriver is a Python library for browser automation that communicates directly with the browser via the Chrome DevTools Protocol (CDP), without relying on Selenium or chromedriver. It is the successor to undetected-chromedriver, created by the same author, and is designed from the ground up to be undetectable by anti-bot systems.

Key advantages over traditional browser automation:

No chromedriver dependency Communicates directly with Chrome/Chromium via CDP websocket, eliminating the most common detection vector (chromedriver fingerprint).
Undetectable by default Does not set the navigator.webdriver flag, does not inject automation-related JavaScript, and avoids CDP detection patterns that anti-bot systems look for.
Fast and lightweight Without the Selenium/WebDriver protocol overhead, nodriver is significantly faster at launching browsers and executing commands.
Async-first Built entirely on Python's asyncio, enabling efficient concurrent browser automation.
Simple API Clean, Pythonic API that is easier to use than raw CDP or Selenium.

nodriver is particularly useful for scraping websites protected by advanced anti-bot systems like Cloudflare, DataDome, or PerimeterX, where standard Selenium or Playwright setups get detected and blocked.

Highlights

ai-powerednatural-languagetypescript

anti-detectcdpasyncfast

Example Use

```javascript import { Stagehand } from '@browserbasehq/stagehand'; import { z } from 'zod'; const stagehand = new Stagehand({ env: 'LOCAL', // or 'BROWSERBASE' for cloud browsers modelName: 'gpt-4o', }); await stagehand.init(); const page = stagehand.page; // Navigate to a page await page.goto('https://news.ycombinator.com'); // Use natural language to interact await page.act('click on the "new" link in the navigation'); // Extract structured data with a schema const stories = await page.extract({ instruction: 'Extract the top 5 story titles and their point counts', schema: z.object({ stories: z.array(z.object({ title: z.string(), points: z.number(), })), }), }); console.log(stories); // Observe available actions on the page const actions = await page.observe('What actions can I take on this page?'); console.log(actions); await stagehand.close(); ```

```python import nodriver as uc import asyncio async def main(): # Launch browser - undetected by default browser = await uc.start() # Open a new tab and navigate tab = await browser.get("https://example.com") # Wait for an element and interact with it search_box = await tab.find("input[name='q']") await search_box.send_keys("web scraping") # Click a button button = await tab.find("button[type='submit']") await button.click() # Wait for navigation and extract content await tab.wait_for("div.results") results = await tab.query_selector_all("div.result") for result in results: title = await result.query_selector("h3") print(await title.get_text()) # Take a screenshot await tab.save_screenshot("results.png") browser.stop() asyncio.run(main()) ```