How does this server differ from standard MCP prompts?
It exposes each prompt template as an MCP *tool* (callable with arguments) rather than as an MCP *prompt* resource, making it easier to use in editors like Cursor and Windsurf.
@gdli6177
基本情報
这是一个基于Model Context Protocol (MCP)的服务器,用于根据用户任务需求提供预设的prompt模板,帮助Cline/Cursor/Windsurf...更高效地执行各种任务。服务器将预设的prompt作为工具(tools)返回,以便在Cursor和Windsurf等编辑器中更好地使用。
ツール一覧はまだありません
この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
MCP Prompt Server is a Model Context Protocol (MCP) server that provides pre‑defined prompt templates as MCP tools. It helps AI agents and editors like Cline, Cursor, and Windsurf execute tasks more efficiently by offering structured prompts for code review, API documentation, code refactoring, and more.
Clone the repository, run npm install inside the prompt-server directory, then start with npm start. The server communicates over stdio. To integrate with Cursor or Windsurf, edit the editor’s MCP configuration file to point to the server’s entry script (src/index.js). The server automatically loads all prompt templates from src/prompts/ on start.
reload_prompts and get_prompt_namessrc/prompts/It exposes each prompt template as an MCP *tool* (callable with arguments) rather than as an MCP *prompt* resource, making it easier to use in editors like Cursor and Windsurf.
Add a YAML or JSON file to the `src/prompts` directory following the required schema (name, description, optional arguments, messages). The server loads new templates on restart, or you can call the `reload_prompts` tool.
Cursor, Windsurf, and Cline (and any MCP client that can connect over stdio transport) are supported. The README provides specific configuration steps for Cursor and Windsurf.
Node.js and npm. The server runs as a Node.js process and communicates via stdio.
The README states that the server will continue to work but will no longer receive feature updates or maintenance. Users are encouraged to migrate to the upgraded platform at [mypromptmcp.com](https://mypromptmcp.com) for ongoing support.
YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context
🚀 The fast, Pythonic way to build MCP servers and clients.

Yuppsy MCP: build and publish real estate websites from your AI assistant. Yuppsy lets your AI assistant create live websites for real estate agents and individual property listings. You don't need a designer or any code
Hindsight: Agent Memory That Learns
Open-source MCP to use Blender with any LLM
コメント