What database types are supported by SQLDB?

SQLDB offers multi-database support, compatible with Microsoft SQL Server (MSSQL), PostgreSQL, and MySQL databases.

What is SQLDB and its primary function?

SQLDB is an MCP (Model Context Protocol) server designed to provide Large Language Models (LLMs) with secure, read-only access to SQL databases such as MSSQL, PostgreSQL, and MySQL.

How does SQLDB ensure the security of my database?

SQLDB strictly enforces read-only access. It uses AST-level SQL parsing with dialect awareness to ensure that only `SELECT` statements are allowed, preventing any data modification operations.

How does SQLDB optimize data for Large Language Models?

It utilizes an LLM-optimized compact columnar result format (column list + value rows), significantly reducing token usage. It also provides pagination and total row count awareness for efficient data handling.

Can SQLDB export large query results?

Yes, SQLDB supports streaming file export of large query results to CSV or JSON formats. This feature is designed for extensive datasets, writing directly to disk without a row-count limit.

SQLDB

Name: SQLDB
Author: akrym1582

byakrym1582

0•

Gestión de Bases de Datos

Herramientas para Desarrolladores

Ciencia de Datos y ML

Provides a read-only Model Context Protocol server, enabling Large Language Models to securely interact with SQL databases.

This tool functions as a read-only Model Context Protocol (MCP) server, specifically engineered to grant Large Language Models (LLMs) secure and controlled access to various SQL databases. It supports MSSQL, PostgreSQL, and MySQL, rigorously enforcing read-only operations via AST-level SQL parsing to prevent data modification. Designed for LLM efficiency, it delivers query results in a compact columnar format to minimize token usage and includes features like pagination, total row counts, and result caching. Additionally, it offers powerful capabilities for streaming large query results to CSV or JSON files.

Características Principales

01Multi-database support for MSSQL, PostgreSQL, and MySQL

02Read-only access enforced via AST-level SQL parsing for security

03LLM-optimised compact columnar result format to reduce token usage

04Pagination and total row count awareness for efficient data handling

05Streaming file export for large query results to CSV or JSON

060 GitHub stars

Casos de Uso

01Empowering LLMs to execute secure, read-only queries against various SQL databases.

02Allowing LLMs to understand database schemas and table structures for informed interactions.

03Enabling LLMs to export extensive database query results to CSV or JSON files for further analysis.