Azure Data Explorer MCP Server

DatabasesPython

Query and analyze Azure Data Explorer databases through standardized interfaces

Available Tools

execute_query

Execute a KQL query against the configured Azure Data Explorer database

query

list_tables

List all tables in the configured database

get_table_schema

Get the schema for a specific table

table_name

sample_table_data

Get sample data from a specific table

table_namelimit

get_table_details

Get detailed statistics and information about a specific table

table_name

Azure Data Explorer MCP Server provides a seamless interface for AI assistants to interact with Azure Data Explorer (ADX) and Eventhouse in Microsoft Fabric. It enables executing KQL queries, exploring database schemas, and analyzing data through standardized Model Context Protocol interfaces. With this MCP, AI assistants can discover database resources, view table schemas, sample data, and get detailed statistics - all while leveraging Azure's authentication mechanisms including token credentials and workload identity support for Kubernetes environments.

Overview

Azure Data Explorer MCP Server enables AI assistants to interact with your Azure Data Explorer (ADX) or Eventhouse databases through the Model Context Protocol. This allows your AI assistant to execute KQL queries, explore database schemas, and analyze your data directly.

Installation

Prerequisites

Access to an Azure Data Explorer cluster
Azure account with appropriate permissions
Python 3.10 or higher

Setup Instructions

Clone the repository:

git clone https://github.com/pab1it0/adx-mcp-server.git
cd adx-mcp-server

Install dependencies:

pip install -e .

Configure environment variables by creating a .env file in the project root:

# Required: Azure Data Explorer configuration
ADX_CLUSTER_URL=https://yourcluster.region.kusto.windows.net
ADX_DATABASE=your_database

# Optional: Azure Workload Identity credentials (for AKS deployments)
# AZURE_TENANT_ID=your-tenant-id
# AZURE_CLIENT_ID=your-client-id
# ADX_TOKEN_FILE_PATH=/var/run/secrets/azure/tokens/azure-identity-token

Authenticate with Azure:
- For local development: Log in using Azure CLI (az login)
- For production: Configure appropriate credentials based on your environment

Authentication Options

The server supports multiple authentication methods:

Default Azure Credential - Automatically tries multiple authentication methods in sequence
Workload Identity Credential - For AKS environments with workload identity configured
Azure CLI Credential - Uses your Azure CLI login

For AKS with Azure Workload Identity, ensure your pod has the required environment variables (AZURE_TENANT_ID and AZURE_CLIENT_ID) and the token file is properly mounted.

Docker Deployment

You can also run the server as a Docker container:

docker build -t adx-mcp-server .
docker run -p 8000:8000 --env-file .env adx-mcp-server

Usage with AI Assistants

Once the server is running, you can configure your AI assistant to use it. The server exposes MCP-compatible endpoints that allow the AI to:

Execute KQL queries against your databases
List available tables
View table schemas
Sample data from tables
Get table statistics and details

The server automatically handles authentication with your Azure Data Explorer cluster using the configured credentials.