executeautomation/mcp-playwright
Mcp Playwright
This project is a Model Context Protocol server that enables Large Language Models to control Playwright browsers for web automation, scraping, and end-to-end testing. It functions as a programmable interface for executing JavaScript, capturing screenshots, and interacting with web elements across multiple browser engines.
The server exposes browser automation capabilities as a set of standardized tools that models can discover and invoke. It supports session-based browser isolation to ensure unique contexts for each client connection and provides a transport layer using either standard input/output or Server-Sent Events.
The system covers broad capability areas including cross-browser testing, device layout emulation, and natural language web testing. It also includes tools for web content extraction, PDF export, and the execution of API requests with custom authentication.
The service can be deployed as a standalone HTTP server to support remote hosting and concurrent connections.
Features
- Browser Automation Tools - Provides a programmable interface for AI agents to navigate, scrape, and interact with web content via Playwright.
- MCP Server Interfaces - Exposes browser automation capabilities as a standardized Model Context Protocol server for AI agent integration.
- Browser Automation Agents - Enables AI agents to interact with web browsers for user interface testing and data scraping.
- MCP Servers - Implements the Model Context Protocol to expose browser automation tools to large language models.
- Content Extraction - Retrieves cleaned, visible text or filtered HTML from web pages by removing scripts and styles.
- DOM-Based Extractions - Retrieves page content by filtering HTML and removing scripts to provide clean text for LLM processing.
- JSON-RPC Implementations - Uses the JSON-RPC protocol over standard I/O or SSE for tool orchestration between client and server.
- Element Interaction - Performs clicks, hovers, drags, and keyboard presses on page elements, including those inside frames.
- Browser Context Isolation - Creates unique browser contexts for each client connection to ensure state isolation.
- Input Simulations - Simulates real user inputs including keyboard presses and drag-and-drop actions to interact with web elements.
- End-to-End Testing Frameworks - Validates complete application workflows by simulating user interactions and monitoring API requests.
- Natural Language Test Generators - Translates plain English instructions into executable browser actions for automated user interface and API testing.
- JavaScript Execution Bridges - Executes custom JavaScript directly within the browser context to manipulate the DOM or extract data.
- Browser Session Managers - Ensures state isolation by creating unique browser instances for every client connection using session IDs.
- Headless Browser Controllers - Utilizes the Playwright framework to orchestrate headless or headed browser instances across multiple engines.
- Browser Navigation - Directs the browser to specific URLs with configurable engines, viewport sizes, and headless modes.
- Web Scraping - Extracts structured content and HTML from websites using a headless browser environment.
- API Request Automation - Enables sending requests to application endpoints and validating responses for automated web service testing.
- Web Page Screenshot Tools - Captures visual snapshots of rendered web pages to verify layout and document interface state.
- Browser Device Emulation - Resizes the browser viewport and configures user-agents and touch settings using device presets.
- Replayable Interaction Recorders - Captures real-time user interactions in the browser and serializes them into test specifications.
- Test Code Generators - Provides a CLI to record browser interactions and generate corresponding test code in a specified directory.
- Viewport and User Agent Simulation - Adjusts viewport dimensions and user-agent strings using predefined profiles to simulate mobile and desktop environments.
- Cross-Engine Rendering Comparisons - Enables opening websites across multiple browser engines to compare rendering and behavior consistency.
- End-to-End Testing - Coordinates interactions across interfaces, APIs, and databases to verify complete system workflows.
- Cross-Browser Testing Tools - Runs automation tasks across multiple browser engines to ensure consistent behavior across environments.
- Input Field Automators - Provides utilities for programmatically entering text, selecting options, and uploading files into web application fields.
- Viewport Simulators - Adjusts browser viewport dimensions to test responsive layouts and simulate various screen sizes.
- Remote Browser Controllers - Provides a standalone server mode allowing remote clients to control browser instances via a network interface.
- Browser Session Managers - Manages browser instances by closing them to release resources and modifying user-agent strings.
- Headless Server Hosting - Runs as a standalone HTTP server to support remote deployments and concurrent client connections.
- Search and Web - Browser automation capabilities using Playwright.
- Browser Automation - Playwright-based web interaction and scraping server.
