
A Model Context Protocol (MCP) server for executing Kusto Query Language (KQL) queries against Azure Data Explorer clusters.
概览
A Model Context Protocol (MCP) server for executing Kusto Query Language (KQL) queries against Azure Data Explorer clusters. Seamlessly integrates with Claude Desktop and Visual Studio Code (VS Code). Supports Azure CLI authentication and provides a convenient kql_execute tool for directly within your , complete with optional visualizations. - : Run KQL queries directly within your AI prompts using the kql_execute MCP tool. - : No manual tenant ID configuration or complex credential management. Simply authenticate via Azure CLI (az login) and start querying immediately. - : Execute KQL queries within Claude Desktop without switching interfaces or dashboards. - : Leverages Azure CLI’s robust authentication, ensuring secure credential handling and optimized query performance. - : Optional Markdown table outputs make data analysis intuitive and engaging. - : Built-in Azure CLI authentication with retry and caching logic. - : Fully compatible with MCP protocol version 2024-11-05.
README
以下为原文 README 节选,完整内容请查看原文。
🚀 MCP KQL Server
A Model Context Protocol (MCP) server for executing Kusto Query Language (KQL) queries against Azure Data Explorer clusters. Seamlessly integrates with Claude Desktop and Visual Studio Code (VS Code). Supports Azure CLI authentication and provides a convenient kql_execute tool for execute KQL queries directly within your AI prompts, complete with optional visualizations.

✨ Features
- 🔍 Execute KQL Queries: Run KQL queries directly within your AI prompts using the
kql_executeMCP tool. - ⚡ Plug-and-Play Simplicity: No manual tenant ID configuration or complex credential management. Simply authenticate via Azure CLI (
az login) and start querying immediately. - 🤖 Seamless Claude Integration: Execute KQL queries within Claude Desktop without switching interfaces or dashboards.
- 🔒 Secure and Efficient: Leverages Azure CLI’s robust authentication, ensuring secure credential handling and optimized query performance.
- 📊 Visualization Ready: Optional Markdown table outputs make data analysis intuitive and engaging.
- 🔑 Azure CLI Authentication: Built-in Azure CLI authentication with retry and caching logic.
- 📡 Protocol Compatibility: Fully compatible with MCP protocol version
2024-11-05. - 🛠️ Reliable Implementation: Uses
fastmcpfor a robust and reliable MCP server implementation.
📌 Comparison with Other Tools
| Feature | MCP KQL Server | Azure Data Explorer MCP | Standalone Scripts |
|---|---|---|---|
| Claude Integration | ✅ | ❌ | ❌ |
| Plug-and-Play Authentication | ✅ (Azure CLI) | ❌ (Manual setup) | ❌ (Custom code) |
| No Tenant ID Required | ✅ | ❌ | ❌ |
| Data Visualization | ✅ | ✅ | ❌ |
| Open-Source | ✅ | ✅ | Varies |
📋 Requirements
- Python 3.10 or higher
- Azure Data Explorer cluster access
- Azure CLI installed and authenticated (
az login) - Node.js (optional, for Claude Desktop filesystem server)
- VS Code with Claude extension or MCP client (for VS Code integration)
📂 Project Structure
mcp-kql-server/
├── mcp_kql_server/
│ ├── __init__.py
│ ├── mcp_server.py
│ ├── kql_auth.py
│ └── execute_kql.py
├── pyproject.toml
├── README.md
🏗️ Setup Instructions
1️⃣ Clone the Repository
git clone https://github.com/4R9UN/mcp-kql-server.git
cd mcp-kql-server
2️⃣ Install Python 3.10+
- Windows:
winget install Python.Python.3.10 - macOS:
brew install [email protected] - Linux:
sudo apt-get install python3.10
安装
This server does not publish a one-line install command.
Open the repository installation guide 新手?先看 mcpskill.org 教程 →配置
{
"mcpServers": {
"mcp-kql-server": {
"command": "C:\\Users\\YourUser\\mcp-kql-server\\.venv\\Scripts\\python.exe",
"args": [
"-m",
"mcp_kql_server.mcp_server"
]
}
}
}