MCP Server for Azure Log Analytics

A Model Context Protocol (MCP) server that exposes Azure Log Analytics workspace data, with specific support for AuditLogs and AzureActivity tables.

Features

• Separate tools for querying AuditLogs and AzureActivity tables

• Custom KQL queries with flexible filtering

• Time range filters with human-readable format (24h, 7d, 30d)

• Pagination support for large result sets

• DefaultAzureCredential authentication (CLI, Managed Identity, Service Principal)

• Docker support for containerized deployment

• Comprehensive unit tests

Prerequisites

• Python 3.9+

• Azure Log Analytics workspace

• Azure CLI (for local development) or Managed Identity (for production)

Installation

1. Clone the repository

2. Create a virtual environment:

   python -m venv venv

3. Activate the virtual environment:

   • Windows (PowerShell): venv\Scripts\Activate.ps1

   • Windows (cmd): venv\Scripts\activate.bat

   • Linux/Mac: source venv/bin/activate

4. Install dependencies:

   pip install -r requirements.txt

Configuration

Set the following environment variables:

• AZURE_LOG_ANALYTICS_WORKSPACE_ID (required): Your Log Analytics workspace ID

• AZURE_TENANT_ID (optional): Azure tenant ID for service principal

• AZURE_CLIENT_ID (optional): Service principal client ID

• AZURE_CLIENT_SECRET (optional): Service principal client secret

Usage

The MCP server communicates via stdio and is designed to be used with MCP clients like VS Code (GitHub Copilot) or Claude Desktop.

Quick Start with VS Code (GitHub Copilot)

1. Create or edit: %APPDATA%\Code\User\globalStorage\github.copilot\mcp-settings.json

   {
     "mcpServers": {
       "azure-log-analytics": {
         "command": "C:\\dev\\sre-agent\\mcp-log-analytics\\venv\\Scripts\\python.exe",
         "args": ["-m", "src.server"],
         "cwd": "C:\\dev\\sre-agent\\mcp-log-analytics",
         "env": {
           "AZURE_LOG_ANALYTICS_WORKSPACE_ID": "your-workspace-id"
         }
       }
     }
   }

2. Restart VS Code

3. Use Copilot Chat to query your logs:

   • @azure-log-analytics Show me audit logs from the last 24 hours

   • @azure-log-analytics Find failed operations in Azure Activity logs this week

For detailed VS Code setup and troubleshooting, see VSCODE-USAGE.md

Development

Install development dependencies:

pip install -r requirements-dev.txt

Run tests:

pytest

Docker

Build the image:

docker build -t mcp-log-analytics .

Run the container:

docker run -e AZURE_LOG_ANALYTICS_WORKSPACE_ID=<workspace-id> mcp-log-analytics

Documentation

• VSCODE-USAGE.md - Complete VS Code & GitHub Copilot setup guide

• DESIGN.md - Architecture and design documentation

• CONTRIBUTING.md - Development and contribution guidelines

License

MIT