Azure Data Explorer MCP

Example Use Cases

• 1

  Execute Kusto Query Language (KQL) queries against an Azure Data Explorer/Eventhouse cluster to fetch or aggregate data.

• 2

  Discover and inspect database resources — list tables, view table schemas, sample rows, and retrieve table statistics.

• 3

  Seamless integration with Azure authentication (including Workload Identity for AKS) and containerized deployment for secure, reproducible access.

Description

This project implements a Model Context Protocol (MCP) server for Azure Data Explorer (ADX) / Eventhouse, allowing AI agents to execute Kusto Query Language (KQL) queries and discover database resources (tables, schemas, samples, and stats). It supports multiple authentication methods including DefaultAzureCredential and WorkloadIdentityCredential for AKS, and can be run over stdio, HTTP/SSE, or inside Docker. The server is configurable so you can choose which tools to expose to MCP clients, and includes devcontainer and Docker support for easy deployment and development.