Ramoira MCP Server
Exposes the Ramoira brand schema specification as MCP tools and resources, enabling LLMs to generate, validate, or reason about brand schemas without the CLI.
README
@ramoira/mcp
MCP server exposing the Ramoira brand schema specification as tools and resources.
Any MCP-compatible client (Claude Desktop, Cursor, Windsurf, etc.) can connect to this server and use it to generate, validate, or reason about Ramoira brand schemas — without needing the CLI installed.
What it exposes
Tools
| Tool | Description |
|---|---|
get_spec_overview |
High-level explanation of the schema format, its five layers, and key concepts. Start here. |
get_layer_spec(layer) |
Full field definitions for one layer: identity, narrative, voice, commercial, or governance. |
get_valid_enums() |
All valid enum values for every constrained field, with field paths. |
get_example_schema() |
A high-quality worked example (Rolex) showing what well-populated fields look like. |
search_spec(query) |
Keyword search across all spec documentation. |
Resources
| URI | Description |
|---|---|
ramoira://spec |
Spec overview as a plain text resource. |
ramoira://enums |
Valid enum values as a JSON resource. |
How it fits into Ramoira
ramoira/brand-schema-spec ← the spec (source of truth)
│
▼
ramoira/mcp ← this package
serves the spec as MCP tools so any LLM can consume it
│
├─► Claude Desktop / Cursor / any MCP client
│ Any LLM session can look up the spec on demand
│
└─► ramoira/cli (internally)
ramoira init uses the same tool set locally
to give Claude spec access during schema generation
The CLI's generator mirrors this tool set locally (no network dependency) so ramoira init works offline. This server is for external LLM workflows — any agent that needs to generate or validate a schema can connect here instead of inlining the spec in its prompt.
Install
npm install -g @ramoira/mcp
Or run directly without installing:
npx @ramoira/mcp
Configure in Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"ramoira": {
"command": "npx",
"args": ["@ramoira/mcp"]
}
}
}
Configure in Cursor / VS Code
Add to your MCP settings:
{
"ramoira": {
"command": "npx",
"args": ["@ramoira/mcp"]
}
}
Public HTTP hosting
The server currently uses stdio transport, which works for local clients. To serve it publicly over HTTP, replace the transport in src/server.ts:
import { StreamableHTTPServerTransport } from "@modelcontextprotocol/sdk/server/streamableHttp.js";
const transport = new StreamableHTTPServerTransport({ port: 3000 });
await server.connect(transport);
Then deploy to any Node.js host. Clients connect to https://mcp.ramoira.com (or wherever it's hosted) via the MCP HTTP client.
Development
npm install
npm run dev # run via tsx (no build step)
npm run build # compile to dist/
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。