
一个模式上下文协议(MCP)服务器,用于协助您完成与开放EHR相关的各种任务和API
概览
一个MCP服务器,帮助AI助手与openEHR的原型,模板,AQL,名词和规格合作. 是为将MCP客户端(Claude Desktop, Cursor, LibreChat,.)与其开放的EHR作品相连接的人. 与开放的EHR合作意味着导航临床知识管理器(CKM),复杂的类型系统,以及ADL语法规则. 服务器通过MCP工具,提示,资源,以及完成等方式,让客户端直接访问这些来源,因此助理可以帮助进行原型探索,语义解释,语言翻译,语法校正,以及设计审查. 服务器拥有MCP表面:工具,提示,和资源,以及CKM访问,向导,示例,名词,以及背后的类型规格. 它不决定助理何时使用。 这个工作流程层是开源EHR助理插件,该插件为克洛德代码和光标添加了技能,命令,代理和钩子;两个配对为导读开源EHR工作流程.
概述(AI 翻译)
一个MCP服务器,帮助AI助手与openEHR的原型,模板,AQL,名词和规格合作. 是为将MCP客户端(Claude Desktop, Cursor, LibreChat,.)与其开放的EHR作品相连接的人. 与开放的EHR合作意味着导航临床知识管理器(CKM),复杂的类型系统,以及ADL语法规则. 服务器通过MCP工具,提示,资源,以及完成等方式,让客户端直接访问这些来源,因此助理可以帮助进行原型探索,语义解释,语言翻译,语法校正,以及设计审查.
服务器拥有MCP表面:工具,提示,和资源,以及CKM访问,向导,示例,名词,以及背后的类型规格. 它不决定助理何时使用。 这个工作流程层是开源EHR助理插件,该插件为克洛德代码和光标添加了技能,命令,代理和钩子;两个配对为导读开源EHR工作流程. Claude代码用户可以从Cadasto插件市场安装插件.
** 要求。 ** 使用[[T0]或stdio运输的MCP客户端.
本页由 AI 翻译,技术细节请以原文为准。
README
以下为原文 README 节选,完整内容请查看原文。
openEHR Assistant MCP Server
An MCP server that helps AI assistants work with openEHR archetypes, templates, AQL, terminology, and specifications. It is for people who connect an MCP client (Claude Desktop, Cursor, LibreChat, …) to their openEHR work. Working with openEHR means navigating the Clinical Knowledge Manager (CKM), intricate type systems, and ADL syntax rules. The server gives the client direct access to those sources through MCP tools, prompts, resources, and completions, so the assistant can help with archetype exploration, semantic explanation, language translation, syntax correction, and design reviews.
The server owns the MCP surface: the tools, prompts, and resources, and the CKM access, guides, examples, terminology, and type specifications behind them. It does not decide when an assistant should use them. That workflow layer is the openEHR Assistant Plugin, which adds skills, commands, agents, and hooks for Claude Code and Cursor; pair the two for guided openEHR workflows. Claude Code users can install the plugin from the Cadasto Plugin Marketplace.
Requirements. An MCP client that speaks the streamable-http or stdio transport. The hosted endpoint needs no install, only network access to https://openehr-assistant-mcp.apps.cadasto.com/. To run your own instance you need Docker with Docker Compose (and Git to clone the repository), or Docker alone to run the published image over stdio. PHP 8.4 ships inside the image, so the host needs no PHP. The CKM tools call the CKM REST API (https://ckm.openehr.org/ckm/rest by default, set with CKM_API_BASE_URL); guides, examples, terminology, and type specifications are bundled with the server.
Pre-release: expect frequent updates and breaking changes until version 1.0.
安装
This server does not publish a one-line install command.
Open the repository installation guide 新手?先看 mcpskill.org 教程 →配置
{
"mcpServers": {
"openehr-assistant-mcp": {
"type": "streamable-http",
"url": "https://openehr-assistant-mcp.apps.cadasto.com/"
}
}
}