Azure Data Explorer MCP Server

What is Azure Data Explorer MCP Server?

An MCP server that enables AI assistants to execute KQL queries and explore Azure Data Explorer (ADX/Kusto) and Eventhouse databases through a standardized interface. It is for developers and analysts who want to query ADX clusters using AI-driven tools.

How to use Azure Data Explorer MCP Server?

Configure the required environment variables (ADX_CLUSTER_URL and ADX_DATABASE), log in to Azure via the Azure CLI, then add the server configuration to your MCP client (e.g., Claude Desktop). You can run it directly with uv, via Docker, or in a Dev Container.

Key features of Azure Data Explorer MCP Server

• Execute arbitrary KQL queries and receive structured JSON results.
• Discover tables, view schemas, sample data, and get table statistics.
• Supports DefaultAzureCredential and Azure Workload Identity authentication.
• Configurable transports: stdio, HTTP, and Server-Sent Events (SSE).
• Docker images and Dev Container support for easy deployment.

Use cases of Azure Data Explorer MCP Server

• AI assistants querying Azure Data Explorer databases on demand.
• Automated data exploration and analysis through natural language.
• Integrating ADX queries into MCP‑compatible clients like Claude Desktop.
• Running ADX queries from CI/CD pipelines or containerized environments.

FAQ from Azure Data Explorer MCP Server

What authentication methods are supported?

The server uses DefaultAzureCredential by default and falls back to WorkloadIdentityCredential when the required environment variables are present (e.g., in AKS with workload identity).

What environment variables are required?

ADX_CLUSTER_URL (the cluster URL) and ADX_DATABASE (the database name) are required.

What Python version is required?

Python 3.12 is required.

Does the server support HTTP transport?

Yes. Set ADX_MCP_SERVER_TRANSPORT to http or sse and optionally configure ADX_MCP_BIND_HOST and ADX_MCP_BIND_PORT.

Can I run this server in Docker?

Yes. A Dockerfile and docker-compose configuration are provided. Pass the required environment variables to the container.