关于
This project serves as an example demonstrating the integration of Playwright with Cloudflare Workers to create a Model Control Protocol (MCP) server. It empowers AI assistants to execute various web automation tasks such as navigation, text input, clicking elements, and capturing screenshots. The server offers a robust set of tools, making it compatible with diverse AI platforms, including Cloudflare AI Playground, Claude Desktop, and GitHub Copilot in VS Code, thus facilitating advanced AI-driven web interactions.
主要功能
- Deployed and operated using Cloudflare Workers for scalability
- Provides AI-driven browser control for web automation tasks
- Supports browser navigation, typing, clicking, and screenshot capture
- Compatible with Cloudflare AI Playground, Claude Desktop, and GitHub Copilot in VS Code
- Exposes 14 distinct browser control tools via the MCP server
- 0 GitHub stars
使用案例
- Integrating browser control capabilities into GitHub Copilot agents within VS Code
- Testing and experimenting with MCP servers using LLMs in Cloudflare AI Playground
- Automating web tasks with AI assistants through Claude Desktop's local MCP proxy