MCP.so
ログイン

Alkemi MCP Server

@alkemi-ai

基本情報

カテゴリ

データベース

トランスポート

stdio

作者

alkemi-ai

投稿者

Matt Ferrante

Alkemi MCP Server について

Integrate your Alkemi Data, connected to Snowflake, Google BigQuery, DataBricks and other sources, with your MCP Client.

Alkemi 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": {
    "alkemi": {
      "command": "npx",
      "args": [
        "@alkemiai/alkemi-mcp"
      ],
      "env": {
        "BEARER_TOKEN": "<YOUR_TOKEN>"
      }
    }
  }
}

Alkemi 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 Alkemi Mcp?

Alkemi Mcp is a STDIO wrapper for the Streamable HTTP MCP endpoint at https://api.alkemi.cloud/mcp that integrates Alkemi Data—connected to Snowflake, Google BigQuery, DataBricks and other sources—with any MCP Client. It uses Alkemi to store database metadata, generate proper queries, and query the database so teams can share consistent data-access capabilities. Get a free API key at datalab.alkemi.ai.

How to use Alkemi Mcp?

Install via npx @alkemiai/alkemi-mcp or build locally from the cloned repository. Add a configuration entry to your Claude Desktop claude_desktop_config.json, setting the required BEARER_TOKEN environment variable. Optionally set MCP_NAME to name the server or PRODUCT_ID to limit scope to a single product. Multiple Alkemi Mcp instances can be configured with different names and product IDs.

Key features of Alkemi Mcp

  • STDIO wrapper for the Streamable HTTP MCP endpoint
  • Stores database metadata for accurate query generation
  • Generates proper SQL queries automatically
  • Queries the underlying database on your behalf
  • Supports multiple database sources (Snowflake, BigQuery, DataBricks, etc.)
  • Allows sharing a single MCP Server across teammates

Use cases of Alkemi Mcp

  • Query databases without requiring users to know table schemas or SQL details
  • Share a single MCP Server so all teammates have the same data-query quality
  • Narrow scope to a specific product or dataset using PRODUCT_ID
  • Connect an MCP client to cloud data warehouses like Snowflake and BigQuery

よくある質問

1

What is Alkemi Mcp and how does it differ from directly querying a database?

Alkemi Mcp uses Alkemi to store database metadata, generate proper queries, and execute them—removing the need to maintain schema information in every agent or MCP Client. This prevents wrong answers and scales better for teams sharing data.

2

How do I get an API key for Alkemi Mcp?

Sign up for a free API key at [datalab.alkemi.ai](https://datalab.alkemi.ai?ref=alkemi-mcp).

3

What environment variables are required?

The `BEARER_TOKEN` is required. `MCP_NAME` and `PRODUCT_ID` are optional (use `MCP_NAME` when configuring multiple instances to avoid duplicate names).

4

What runtime or dependencies are needed?

Alkemi Mcp runs via Node.js. It can be executed with `npx` (no local install) or built from the repository using `npm install && npm run build`. The server communicates over stdio.

5

How does authentication work for Alkemi Mcp?

Authentication is handled by setting the `BEARER_TOKEN` environment variable, which provides the bearer token for the underlying Streamable HTTP MCP endpoint.

コメント

「データベース」の他のコンテンツ