punkpeye/awesome-remote-mcp-servers

What it solves

This project provides a curated directory of remote Model Context Protocol (MCP) servers. Unlike traditional MCP servers that require you to install software, manage local processes, and handle updates yourself, these servers are hosted by providers and accessed directly via a URL.

How it works

Remote MCP servers move the infrastructure burden to the provider. You connect your AI client (such as Claude Code, Cursor, or VS Code) to a specific endpoint URL. Depending on the server, you may need to authenticate using an API key or an OAuth browser flow. The server handles the execution, and you simply interact with the tools it exposes.

Who it’s for

This list is for developers and AI power users who want to extend their AI agents with external capabilities—like web search, data analysis, or third-party app integrations—without the complexity of running and maintaining local server processes.

Highlights

• No local installation: Access powerful tools without managing packages, binaries, or Docker containers.
• Managed infrastructure: Providers handle updates and maintenance, ensuring the server is always running.
• Diverse authentication: Includes options for anonymous access (🔓), API keys (🔑), and secure OAuth (🔐).
• Verified quality: Many entries feature a connector badge, indicating they have been independently scored for endpoint health and tool quality.
• Extensive categories: Covers a wide range of use cases including cloud platforms, databases, developer tools, and project management.