# Hyperbrowser Hyperbrowser provides on-demand cloud browsers and sandboxes for AI agents, web automation, scraping, testing, and agentic code execution. It connects browser workflows through an API, Playwright, Puppeteer, Python, or Node.js. - Findings24: - ウェブサイト: - 料金: Freemium - カテゴリー: AIエージェント基盤, AIブラウザ自動化, AIウェブスクレイパー ## Hyperbrowserとは？ Hyperbrowser is browser infrastructure for AI agents and automation. It provisions cloud browser sessions on demand and lets existing Playwright or Puppeteer scripts connect through a browser session, so teams can run web workflows without managing browser infrastructure directly. The documentation also presents sandboxes for agentic code execution and large-scale web scraping. The platform supports browser session management, Web API access, AI-powered browser automation, stealth and proxy configuration, and SDK-based integration. Its home-page example shows a Python workflow that creates a session, connects Playwright over CDP, navigates to a flight-search site, fills form fields, and interacts with the page. Hyperbrowser uses credit-based billing. The pricing page lists a free plan with 5,000 credits, one concurrent browser, and seven days of data retention; paid plans add higher concurrency and features such as captcha solving, stealth modes, and residential proxies. Enterprise availability includes higher concurrency, longer retention, premium support, and compliance offerings stated on the pricing page. ## Hyperbrowserでできること ### On-demand cloud browser sessions Create and manage browser sessions through the platform instead of running browser infrastructure locally. ### Playwright and Puppeteer connectivity Connect existing automation scripts to cloud browsers, including a documented Playwright-over-CDP workflow. ### Web API and SDK access Use the Web API or the documented Python and Node SDKs to integrate browser sessions into applications. ### AI-agent and sandbox workflows Support AI-powered browser automation and sandbox-based agentic code execution alongside scraping and automation tasks. ### Stealth and proxy configuration Documentation covers anti-detection features and proxy configuration; paid plans list basic or ultra stealth modes and residential proxy access at applicable tiers. ### Usage-based scaling Credit-based billing measures active browser time and proxy data, while paid plans provide higher concurrent-browser limits. ## 利用シーン ### Web research and data collection Run cloud browser sessions for web scraping workflows that need to navigate pages and interact with sites rather than only issue basic requests. ### Browser-based AI agents Give an AI-powered automation agent a managed browser session for tasks such as navigating websites, completing forms, and extracting results. ### Automated testing Connect Playwright or Puppeteer test scripts to remote cloud browsers so browser execution is handled as an external service. ### Agentic code execution Use the documented sandbox capability for workflows in which an agent needs an execution environment in addition to browser access. ### Interactive web automation Automate multi-step processes such as the documented flight-search example, including navigation, field entry, and button interaction. ## よくある質問 ### How do I connect an existing browser automation script? Use a Hyperbrowser session and connect an existing Playwright or Puppeteer workflow to the cloud browser. The home-page example uses Playwright's \`connect\_over\_cdp\` with a session WebSocket endpoint. ### Does Hyperbrowser provide SDKs? The documentation includes Python and Node SDK resources, as well as a Web API and API reference. ### What does Hyperbrowser usage cost? The pricing page uses credits: active browser sessions are billed at 100 credits per hour per browser instance, proxy usage at 10,000 credits per gigabyte, and API calls are listed as free. Plans and included credits vary by tier. ### Is there a free plan? Yes. The pricing page lists a free plan with 5,000 included credits, one concurrent browser, and seven days of data retention. ### Can sessions use stealth or residential proxies? The documentation includes stealth and proxy configuration resources. The pricing page lists basic or ultra stealth modes and residential proxy access on specified paid tiers. ## クイック情報 - **Category**: Browser infrastructure for AI agents and automation - **Primary workflows**: Web scraping, browser automation, testing, and agentic code execution - **Integration methods**: Web API, Python SDK, Node SDK, Playwright, and Puppeteer - **Billing model**: Credit-based usage - **Free tier**: 5,000 credits; one concurrent browser; seven days of data retention - **Pricing domain**: hyperbrowser.ai