mcp-server-mariadb-vector

mcp-server-mariadb-vector

0
databases

The MariaDB Vector MCP server provides a natural language interface to interact with a MariaDB database, enhancing LLM agent capabilities through vector and document management. It integrates well with various applications and uses OpenAI's embedding models for efficient semantic searches.

MCP Server MariaDB Vector

The MariaDB Vector MCP server facilitates interaction with a MariaDB database using natural language, allowing seamless integration with LLM agents. It offers various management features for vector stores and documents, and uses OpenAI's embedding models for semantic searches. Compatible with any MCP client, the server is designed to integrate easily into applications like Claude Desktop and Windsurf.

Features

  • Vector Store Management: Create, delete, and list vector stores.
  • Document Management: Add and query documents with metadata.
  • Embedding Provider: Utilize OpenAI's embedding models.

Setup

  1. Clone the repository.
  2. Run as a Python package using uv or as a Docker container.

Configuration Environment Variables

  • MARIADB_HOST, MARIADB_PORT, MARIADB_USER, MARIADB_PASSWORD, MARIADB_DATABASE
  • EMBEDDING_PROVIDER, EMBEDDING_MODEL
  • OPENAI_API_KEY

Integration

Integrate with Claude Desktop, Cursor, or Windsurf using stdio or SSE transport.