# Browser automation

A Playwright-based browser automation MCP server that provides web page interaction capabilities for LLMs.

The Browserbase MCP Server is an open - source project based on the Model Context Protocol (MCP), providing cloud browser automation capabilities, supporting web page interaction, data extraction, screenshot taking, and JavaScript execution, etc., to help LLM applications seamlessly integrate with external data sources and tools.

An MCP server that provides browser automation capabilities for AI models, enabling models such as Claude and Gemini to perform web browsing, searching, and interaction operations through natural language instructions.

A penetration testing tool that automatically detects XSS and SQL injection vulnerabilities in web pages, providing browser interaction, screenshot, and script execution functions

An MCP server based on Playwright, providing browser automation capabilities for LLMs, including web page interaction, screenshot capture, and JavaScript execution.

A browser automation MCP server based on Deno 2 and Playwright, providing web page interaction and screenshot capabilities

An MCP server that enables interaction with the OneNote web version through browser automation, supporting AI assistants to programmatically operate shared notebooks.

The Browserbase MCP Server is an open - source project based on the Model Context Protocol (MCP), providing cloud browser automation services and supporting functions such as web page interaction, data extraction, and JavaScript execution.

MCP-Undetected-Chromedriver is an MCP service based on undetected-chromedriver, providing browser automation control interfaces that bypass anti-scraping detection. It is suitable for web page automation testing and data collection scenarios.

The MCP server provides an API service for browser automation through natural language commands, combining FastMCP, browser - use, and OpenAI technologies.

The Safari MCP server is a Model Context Protocol service that provides automation and developer tool access for Safari browsers, supporting control of Safari sessions, retrieval of console logs, network activity monitoring, and page automation operations.

An MCP server based on Puppeteer, providing browser automation capabilities, supporting X11 and Wayland display servers on Linux, including interaction tools such as page navigation, screenshot taking, and clicking.

The Crawl4AI MCP Server is a Python-based web crawler project that integrates Playwright browser automation, the DuckDuckGo search engine, and the MCP protocol SDK for intelligent web data collection.

Minion Works is a modular AI agent framework that connects to browsers to autonomously perform complex tasks, suitable for developers, researchers, and creative builders.

An MCP server based on Playwright, providing browser automation capabilities for LLMs

An optimized Playwright MCP server that uses a client-server architecture to achieve browser automation. Through an innovative semantic snapshot algorithm, it significantly reduces the amount of page content transmission, while retaining key elements and reducing token consumption by 90%.

An MCP server based on Playwright that provides browser automation capabilities for LLMs, supporting web page interaction, screenshotting, test code generation, and JavaScript execution.

YetiBrowser MCP is a fully open - source implementation of browser automation MCP. It connects through a local MCP server and a browser extension, enabling AI assistants to control real browser tabs. All data is processed locally to ensure transparency and security.

BrowserControl is an MCP server that provides real browser automation capabilities for AI agents. It uses a vision - first approach, enabling interactions such as clicking and inputting through numbered elements, without the need for CSS selectors or XPath.

A server based on the MCP protocol that drives local Chrome/Chromium browser tabs through the Chrome DevTools Protocol (CDP) to achieve automated functions such as clicking, navigating, and taking screenshots without opening the remote debugging port.

Codify MCP is an MCP server based on Apify Agent that can convert browser automation tasks into tools that AI assistants can call, supporting direct passing of tool definitions through JSON parameters without the need for configuration files.

The MCP server for AdsPower browser's LocalAPI, supporting browser creation and configuration operations through LLM instructions.

The MCP Browser Automation Server is a browser control tool based on REST API, supporting functions such as creating sessions, page navigation, taking screenshots, element operations, and real-time console log monitoring.

This project is a Python-based multi-MCP server management framework that interacts with the Google Gemini API through the SSE protocol. It includes two major function modules: calendar management and browser automation, and supports dynamically loading multiple services through a configuration file.

A TypeScript-based MCP protocol server supporting file operations, Shell commands, and browser automation.

A modern browser application based on React that provides a user-friendly interface to interact with the Model Context Protocol (MCP) server in real-time through Server-Sent Events (SSE). It supports features such as theme switching, screenshot preview, and message history.

nova-act-mcp-server is a zero-install MCP protocol server that provides the Amazon Nova Act browser automation tool.

A browser automation tool based on Playwright, supporting the Greasemonkey API and Chrome debugging functions

AI-driven browser automation server based on the MCP protocol, supporting multiple LLM providers and visual interaction

An MCP server based on Playwright that provides browser automation capabilities for LLMs