MCP.so
ログイン

Azure Data Explorer MCP Server

@MCP-Mirror

Azure Data Explorer MCP Server について

Mirror of

基本情報

カテゴリ

データと分析

ライセンス

MIT license

ランタイム

python

トランスポート

stdio

公開者

MCP-Mirror

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": {
    "pab1it0_adx-mcp-server": {
      "command": "docker",
      "args": [
        "build",
        "-t",
        "adx-mcp-server",
        "."
      ]
    }
  }
}

ツール

4

Query

Discovery

Discovery

Discovery

概要

What is Azure Data Explorer MCP Server?

Azure Data Explorer MCP Server is a Model Context Protocol (MCP) server that enables AI assistants to execute KQL queries against Azure Data Explorer clusters and explore database resources. It provides standardized MCP interfaces for data discovery and querying.

How to use Azure Data Explorer MCP Server?

Log in to Azure via the Azure CLI, then set environment variables ADX_CLUSTER_URL and ADX_DATABASE. Add the server configuration to your MCP client (e.g., Claude Desktop) using the uv command with the provided JSON example. Optionally, build and run the Docker image for containerized deployment.

Key features of Azure Data Explorer MCP Server

  • Execute KQL queries against Azure Data Explorer.
  • List tables in the configured database.
  • View table schemas and sample data.
  • Token credential authentication (Azure CLI, MSI, etc.).
  • Configurable tool selection to limit context window usage.
  • Docker containerization support.

Use cases of Azure Data Explorer MCP Server

  • Let an AI assistant answer questions by running live KQL queries.
  • Quickly discover table schemas and sample data in ADX databases.
  • Automate data exploration workflows through MCP client integration.
  • Provide controlled, context-efficient access to ADX for LLM tools.

FAQ from Azure Data Explorer MCP Server

How do I authenticate to Azure Data Explorer?

Login to your Azure account using the Azure CLI before starting the server. Token credential support includes Azure CLI and Managed Service Identity (MSI).

What environment variables are required?

Two variables must be set: ADX_CLUSTER_URL (e.g., https://yourcluster.region.kusto.windows.net) and ADX_DATABASE (your database name).

Can I deploy this server using Docker?

Yes. A Dockerfile and docker-compose.yml are provided. Build the image with docker build -t adx-mcp-server . and run with the required environment variables.

Are the available tools configurable?

Yes, the list of tools is configurable so you can enable only the functionality you need and reduce context window usage.

What license does this project use?

The project is licensed under the MIT License.

コメント

「データと分析」の他のコンテンツ