adx-mcp-serverA Model Context Protocol (MCP) server that enables AI assistants to query and analyze Azure Data Explorer databases through standardized interfaces.

— (0)

Reviewed by Daniel Nikulshyn·Updated July 2026

Overview

The adx-mcp-server is a Model Context Protocol (MCP) server that enables AI assistants to query and analyze Azure Data Explorer (ADX) databases through standardized interfaces. It allows AI assistants to execute KQL queries and explore ADX/Kusto databases, providing seamless access to Azure Data Explorer and Eventhouse clusters in Microsoft Fabric. The server supports multiple authentication methods, including Azure CLI, Managed Identity, and Workload Identity, and offers flexible deployment options, such as Docker and Server-Sent Events (SSE).

Key features

Query Execution: Execute KQL queries against ADX database
Structured results: Get results formatted as JSON
Database Discovery: List tables, View schemas, Sample data, Table statistics
Authentication: Supports Azure CLI, Managed Identity, Workload Identity, and more
Deployment Options: Multiple transports (stdio, HTTP, SSE), Docker support, Dev Container

Pricing

Model: Free
Category: MCP Servers
Rating: No reviews yet

Use cases

AI-Powered Data Analysis

Use the adx-mcp-server to enable AI assistants to analyze data in Azure Data Explorer databases, providing insights and recommendations.

Automated Data Exploration

Utilize the adx-mcp-server to automate data exploration and discovery in Azure Data Explorer databases, streamlining data analysis and decision-making processes.

Pros & Cons

Pros

Enables AI assistants to query and analyze Azure Data Explorer databases through standardized interfaces
Supports multiple authentication methods, including Azure CLI, Managed Identity, and Workload Identity
Offers flexible deployment options, such as Docker and Server-Sent Events (SSE)

Cons

Requires configuration of environment variables for ADX cluster and Azure authentication
May require additional setup for non-stdio transports

Reviews

No reviews yet. Be the first!

Q&A

What configuration is required before running the server?

You need to set environment variables for your ADX cluster (ADX_CLUSTER_URL, ADX_DATABASE) and optionally Azure Workload Identity credentials (AZURE_TENANT_ID, AZURE_CLIENT_ID, ADX_TOKEN_FILE_PATH). These can be set via a .env file or system environment variables.

Asked by Wei Chen · Oct 17, 2025

Can the server return query results in a format usable by other tools?

Yes, query results are returned as structured JSON, making it easy to consume with other applications or AI assistants. It also offers database discovery features like table lists, schema views, sample data, and statistics.

Asked by Katarzyna Zielinska · Sep 21, 2025

How can I deploy adx-mcp-server for production?

The server can be run in Docker with production‑ready images and supports multiple transports such as stdio, HTTP, and Server‑Sent Events (SSE). A Dev Container is also available for a seamless development experience.

Asked by Nadia Benali · Sep 9, 2025

What authentication options does adx-mcp-server support?

It supports Azure CLI, Managed Identity, and Workload Identity via DefaultAzureCredential, plus additional Azure authentication methods. You can also provide custom tokens through environment variables.

Asked by Quang Nguyen · Aug 26, 2025