Kusto MCP

🚀 Natural Language to Kusto Query MCP Server

A configurable MCP server that can convert natural language into Kusto queries and execute these queries on Kusto.

🚀 Quick Start

This is a configurable MCP server that can convert natural language into Kusto queries and execute these queries on Kusto.

✨ Features

• Configuration: Can be configured via a YAML file.
• Query Generation: Generate Kusto queries based on prompts.
• Query Execution: Execute the generated Kusto queries.
• Multi-database Support: Support multiple Kusto databases.

🛠️ Tool List

1. Configuration Tool

• Purpose: Configure via a YAML file.
• Input Parameters:
  • endpoint: Specify the endpoint of the Kusto cluster.
  • appid and appkey: Credentials for authentication.

2. Query Generation Tool

• Purpose: Generate Kusto queries based on prompts.
• Input Parameters:
  • prompt: Prompt information provided by the user.
  • context (optional): Context information provided in JSON format.
  • options (optional): Option information also provided in JSON format.

3. Query Execution Tool

• Purpose: Execute Kusto queries and return results.
• Input Parameters:
  • query: Query statement to be executed.
  • context (optional): Context information provided in JSON format.
  • options (optional): Option information also provided in JSON format.

4. Multi-database Support Tool

• Purpose: Support switching between multiple Kusto databases.
• Input Parameters:
  • database: Specify the name of the database to operate on.

💻 Usage Examples

Referencing Environment Variables

In the configuration file, you can reference the following environment variables:

{
  "endpoint": "your-kusto-endpoint",
  "appid": "${env:AZURE_CLIENT_ID}",
  "appkey": "${env:AZURE_CLIENT_SECRET}"
}

Model Configuration Example

{
  "model": {
    "type": "openai",
    "config": {
      "temperature": 0.7,
      "max_tokens": 1000,
      "api_key": "${env:AZURE_OPENAI_KEY}"
    }
  },
  "kusto": {
    "clusters": {
      "your-cluster-name": {
        "endpoint": "https://your-kusto-endpoint",
        "appid": "${env:AZURE_CLIENT_ID}",
        "appkey": "${env:AZURE_CLIENT_SECRET}"
      }
    }
  }
}

Kusto Configuration Example

clusters:
  your-cluster-name:
    endpoint: https://your-kusto-endpoint
    auth:
      appid: your-app-id
      appkey: your-app-key

Usage in VS Code

Add the following content to .vscode/mcp.json:

{
  "inputs": [
    {
        "type": "promptString",
        "id": "azure-open-ai-key",
        "description": "Please enter your Azure OpenAI key",
        "password": true
    },
    {
        "type": "promptString",
        "id": "kusto-token",
        "description": "Please enter your Kusto token",
        "password": true
    }
  ],
  "kusto": {
    "type": "stdio",
    "command": "docker",
    "args": [
        "run",
        "-i",
        "--rm",
        "-v",
        "/path/to/settings.yaml:/app/settings.yaml", // Enter the path to the settings.yaml file. You can use VSCode variables.
        // Include environment variables referenced in the settings.yaml file
        "-e",
        "AZURE_OPENAI_KEY",
        "-e",
        "KUSTO_ACCESS_TOKEN",
        "alexeyler/kusto-mcp-server"
    ],
    "env": {
        "AZURE_OPENAI_KEY": "${input:azure-open-ai-key}",
        "KUSTO_ACCESS_TOKEN": "${input:kusto-token}"
    }
  }
}

🛠️ Build Instructions

Run the following command from the project root directory:

docker build ./src/Server/Dockerfile src

⚠️ Important Note

• Ensure that all endpoint URLs are correct and accessible.
• Update the prompts section to include relevant queries and instructions. The more detailed, the better.
• Use the provided example configuration as a template to create your own settings.yaml file.