Puppeteer Browser Automation Library

A Node.js library developed by the Google Chrome DevTools team, currently at version 25.2.0 with over 89,000 GitHub stars, providing a high-level API to programmatically control Chrome, Chromium, and Firefox browsers through the DevTools Protocol and the WebDriver BiDi specification. The library runs headless by default with no visible browser UI, making it ideal for automated testing, web scraping, screenshot generation, and PDF rendering in serverless and CI/CD environments. Installation via npm downloads a compatible Chromium binary automatically, while the puppeteer-core package allows connecting to existing browser installations without downloading. Core capabilities include page navigation with network idle detection, element interaction through CSS selectors and ARIA locators with built-in auto-waiting, keyboard and mouse input simulation, cookie and localStorage management, HTTP request interception and response mocking, file upload and download handling, dialog management, geolocation override, viewport and device emulation for responsive testing, and full-page and element-level screenshot capture. A chrome-devtools-mcp server exposes browser automation capabilities as Model Context Protocol tools for AI coding assistants including Claude Code, Cursor, and GitHub Copilot, with an experimental WebMCP API for deeper agent integration. The codebase is maintained at github.com/puppeteer/puppeteer under the Apache 2.0 license, with npm weekly downloads exceeding 5 million and adoption across companies including Google, Microsoft, and Mozilla.