vectordb-mcp-server

bui21x/vectordb-mcp-server

3.1

If you are the rightful owner of vectordb-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

A Model Context Protocol (MCP) server that provides Vector Database integration (Pinecone) for AI agents.

Vector DB MCP Server

A Model Context Protocol (MCP) server that provides Vector Database integration (Pinecone) for AI agents.

Features

  • Vector storage and retrieval
  • Semantic search capabilities
  • Namespace support
  • Metadata storage
  • Health monitoring

Setup

  1. Install dependencies:
pip install -r requirements.txt
  1. Set environment variables:
PINECONE_API_KEY=your_api_key
PINECONE_ENV=your_environment
PINECONE_INDEX=your_index_name
  1. Run server:
uvicorn src.mcp_server:app --reload

API Endpoints

  • POST /upsert - Store vectors
  • POST /search - Search vectors
  • GET /health - Check server health

Example Usage

# Store vectors
POST /upsert
{
    "texts": ["example text 1", "example text 2"],
    "namespace": "my-namespace",
    "metadata": {"key": "value"}
}

# Search vectors
POST /search
{
    "query": "search query",
    "namespace": "my-namespace",
    "top_k": 5
}

MCP Integration

This server follows the MCP specification for tool integration with AI agents.