semantic-postgres-mcp

semantic-postgres-mcp

0
databasesai_chatbot

The Semantic PostgreSQL MCP Server allows AI assistants like Claude to conduct semantic searches on PostgreSQL databases by understanding database structures and executing natural language queries. It leverages vector embeddings to provide context-aware, hybrid search capabilities without altering existing database schemas.

Semantic PostgreSQL MCP Server

A Model Context Protocol (MCP) server enabling AI assistants like Claude to perform semantic search on PostgreSQL databases.

Key Features

  • Vector-based semantic search using OpenAI's embeddings.
  • Metadata-driven storage in vector format.
  • No schema modification necessary.
  • Simple integration with Claude Desktop.

Prerequisites

  • PostgreSQL 13+ with pgvector extension.
  • Node.js 18+.
  • OpenAI API key.
  • Claude Desktop.

Usage Examples

  • Query customer information or product stock levels.
  • Find orders above a specific amount.

How It Works

  1. Stores metadata with vector embeddings.
  2. Converts questions to embeddings to match with schema.
  3. Generates and executes SQL queries.
  4. Presents formatted results.

Installation

  1. Set up the database with the schema.sql script.
  2. Install and configure the MCP server.
  3. Configure Claude Desktop with the MCP server.