home / mcp / azure tablestore mcp server

Azure TableStore MCP Server

A MCP Server to query a Azure Table Storage for local development

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "dkmaker-mcp-azure-tablestorage": {
      "command": "node",
      "args": [
        "C:/path/to/your/mcp-azure-tablestorage/build/index.js"
      ],
      "env": {
        "AZURE_STORAGE_CONNECTION_STRING": "your_connection_string_here"
      }
    }
  }
}

You can run an MCP server that lets you interact with Azure Table Storage directly from your MCP client. It enables querying tables with filters, inspecting table schemas, and listing all tables in a storage account, all while handling errors clearly and requiring only a connection string to connect to Azure Storage.

How to use

Once you have the Azure TableStore MCP server running, you can perform practical data operations from your MCP client. You can query a table with an OData-style filter, retrieve the table schema to understand data structure, and list every table in your storage account. The query tool returns a total count of matching items and a safe, limited subset of items (default is 5) to keep the context window manageable. Do not increase this limit unless you have explicit user confirmation.

How to install

Prerequisites you need before installing the server:

Install Node.js and npm on your system. You can verify installation with the following commands.

Clone the repository and move into the project directory.

# Local development setup
git clone https://github.com/dkmaker/mcp-azure-tablestorage.git
cd mcp-azure-tablestorage
npm install
npm run build

If you prefer to install the package globally and run it directly, you can install and run it with npm or npx.

# Global installation
npm install -g dkmaker-mcp-server-tablestore

# Or run directly with npx
npx dkmaker-mcp-server-tablestore

Configuring the MCP server for Cline

When using Cline, configure the MCP server in your global settings so you can launch it as part of your workflow. The example below shows how to point to your built server and provide the required Azure Storage connection string.

{
  "mcpServers": {
    "tablestore": {
      "command": "node",
      "args": ["C:/path/to/your/mcp-azure-tablestorage/build/index.js"],
      "env": {
        "AZURE_STORAGE_CONNECTION_STRING": "your_connection_string_here"  // Required: Your Azure Storage connection string
      }
    }
  }
}

Available tools

query_table

Query a table using an OData-like filter and return a total matching count along with a safe, limited subset of results (default 5 items) for LLM processing.

get_table_schema

Fetch and display the schema for a specified table to understand its data structure.

list_tables

List all tables present in the configured Azure Storage account.