mcp-odbc-server

mcp-odbc-server

4
databases

The project is an MCP ODBC server allowing seamless interaction between Large Language Models and ODBC-compatible data sources. It operates using a TypeScript layer over node-odbc for efficient data access and query execution.

Introduction

This document provides setup and usage instructions for an MCP ODBC server, designed to give Large Language Models access to ODBC data sources. The server uses a Data Source Name configured for a specific ODBC driver.

Server Implementation

The MCP Server, developed in TypeScript, utilizes node-odbc to manage communications with the ODBC Driver Manager, primarily through node.js.

Usage Overview

  • get_schemas: Lists accessible database schemas.
  • get_tables: Lists tables in a specific schema.
  • describe_table: Provides table details like column names and data types.
  • query_database: Executes SQL queries returning results in JSON or JSONL formats.
  • sparql_query: Executes SPARQL queries, specific to Virtuoso.

Installation

  1. Clone the repository.
  2. Install necessary MCP components and ODBC drivers.
  3. Set configuration files, such as .env for environment variables.