MCP as a Service is a Fair-code platform that allows you to easily deploy MCP servers that provide tools to AI agents, and that are reachable from anywhere, even cloud solutions like N8N — while still being available only to you and your AI workflows thanks to strong authentication.
The project is currently in an early stage of development, but already provides a fully functional "Playwright MCP in the cloud" solution.
Using our hosted service, you can deploy such an MCP server with a single click, and connect it to your AI agents with minimal setup. This in turn allows your AI agents to remote-control a real Google Chrome browser, and surf the web like a human would.
The software powering our hosted service is provided using the Fair-code model, and its code is available on GitHub.
MCP as a Service is a Fair-code project.
Clone the code, run your own instance, or help us build the future of MCP server management.
All contributions are welcome!
Let's break down how Playwright MCP Cloud enables AI agents to interact with real browsers.
With a single click, deploy a full Playwright MCP Server instance in the cloud. We handle all the setup, including:
⚡ Launching an MCP Server takes less than 10 seconds.
Connect your AI agent to the MCP Server using the provided endpoint and authentication token. The server is:
Watch and interact with the browser session through our integrated VNC client:
Our MCP Server integrates seamlessly with various AI platforms and tools:
Deploy your first MCP Server instance and start enabling your AI agents to interact with real browsers.
⚡ Launching an MCP Server takes less than 10 seconds.
Already signed up? Manage your MCP instances here.