広告掲載
MCP.so
ログイン

Ragie MCP Server

@andyciggy

リポジトリ

基本情報

カテゴリ

その他

ライセンス

MIT license

トランスポート

stdio

作者

andyciggy

Ragie MCP Server について

概要はまだありません

#ragie-mcp#model-context-protocol#knowledge-retrieval

Ragie MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "ragie-mcp-server-andyciggy": {
      "command": "npx",
      "args": [
        "@ragieai/mcp-server"
      ],
      "env": {
        "RAGIE_API_KEY": "your_api_key"
      }
    }
  }
}

Ragie MCP Server のツール

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。

上のボタンで一時的なサンドボックスでこのサーバーを実行し、ツール一覧を取得できます。

概要

What is Ragie Model Context Protocol Server?

The Ragie Model Context Protocol Server implements the Model Context Protocol (MCP) to enable AI models to retrieve information from a Ragie knowledge base. It provides a single tool called "retrieve" that allows querying the knowledge base for relevant information. It is built for developers and AI assistants that need to integrate Ragie’s knowledge retrieval capabilities.

How to use Ragie Model Context Protocol Server?

Install and run with npx, setting the required RAGIE_API_KEY environment variable. Optionally pass --description or --partition command line flags. The server listens on stdio for MCP messages and can be configured in Cursor via a mcp.json file or shell script, or in Claude Desktop via claude_desktop_config.json.

Key features of Ragie Model Context Protocol Server

  • Single retrieve tool for querying the knowledge base.
  • Parameters: query (string), topK (optional, default 8), rerank (optional, default true), recencyBias (optional, default false).
  • Returns an array of content chunks with matching text.
  • Supports custom tool descriptions via --description flag.
  • Supports partition-scoped queries via --partition flag.
  • Compatible with Cursor and Claude Desktop MCP clients.

Use cases of Ragie Model Context Protocol Server

  • Enable AI assistants to search a company knowledge base for answers.
  • Provide relevant document snippets to language models during conversations.
  • Integrate Ragie retrieval into automated workflows that use MCP-compatible tools.
  • Offer context-aware responses in Cursor or Claude Desktop environments.

よくある質問

1

What is the `retrieve` tool and what parameters does it accept?

The `retrieve` tool searches the Ragie knowledge base. It accepts `query` (required), `topK` (optional, default 8), `rerank` (optional, default true), and `recencyBias` (optional, default false). It returns an array of content chunks with matching text.

2

What are the prerequisites and dependencies?

Node.js >= 18 and a valid Ragie API key. The server uses `@modelcontextprotocol/sdk`, `ragie`, and `zod`.

3

How do I configure the server with Cursor?

Create a `mcp.json` file in the project’s `.cursor` folder or in `~/.cursor/` with the command `npx -y @ragieai/mcp-server` and required environment variables. Alternatively, use a shell script and add it through Cursor Settings > MCP Servers.

4

How do I configure the server with Claude Desktop?

Add an entry to `claude_desktop_config.json` (located at `~/Library/Application Support/Claude/` on macOS or `%APPDATA%/Claude/` on Windows) with the same command and environment variables, then restart Claude desktop.

5

Are there any command line options or customizations?

Yes: `--description` or `-d` to override the default tool description, and `--partition` or `-p` to specify a Ragie partition ID. Both can be used together.

コメント

「その他」の他のコンテンツ

DeepMark

DeepMark helps teachers deliver rapid, consistent marking with meaningful feedback for every student — in a fraction of the time. What once took a week, now takes one free period.

ICSS

chokcoco

不止于 CSS

20.7K

Volley.sh

Volley

A good sequence diagram can solve almost any communication problem in engineering. Volley lets your agent write them as plain text, with the real request and response payloads on every arrow. It checks its own syntax wi

EverArt

modelcontextprotocol

Model Context Protocol Servers

88.1K

Glasswarp

Glasswarp

See and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.

広告掲載