Pinecone Vector Db MCP Server

🚀 MCP Pinecone Vector Database Server

This project implements a Model Context Protocol (MCP) server that allows reading from and writing to a Pinecone vector database. It is suitable for handling PDF data and Confluence data.

🚀 Quick Start

The MCP Pinecone Vector Database Server enables seamless interaction with the Pinecone vector database, supporting operations such as searching for similar documents, adding new vectors, and processing Confluence data.

✨ Features

• Search for similar documents using text queries.
• Add new vectors to the database with custom metadata.
• Batch process and upload Confluence data.
• Delete vectors by ID.
• Retrieve basic database statistics (currently unavailable).

📦 Installation

Prerequisites

• Bun runtime environment
• Pinecone API key
• OpenAI API key (for generating embeddings)

Steps

1. Clone this repository.
2. Install dependencies:

   bun install
   

3. Create a .env file with the following content:

   PINECONE_API_KEY=Your Pinecone API key
   OPENAI_API_KEY=Your OpenAI API key
   PINECONE_HOST=Your Pinecone host address
   PINECONE_INDEX_NAME=Your index name
   DEFAULT_NAMESPACE=Default namespace
   

💻 Usage Examples

Running the MCP Server

Start the server:

bun src/index.ts

The server will start and listen for MCP commands on stdio.

Running the Example Client

Test the server using the example client:

bun examples/client.ts

Processing Confluence Data

The Confluence processing script provides detailed logging and validation:

bun src/scripts/process-confluence.ts <file path> [collection] [scope]

Parameters:

• file path: Path to your Confluence JSON file (required).
• collection: Document collection name (default is "documentation").
• scope: Document scope (default is "documentation").

Example:

bun src/scripts/process-confluence.ts ./data/confluence-export.json "tech-docs" "engineering"

The script will perform the following actions:

1. Validate input parameters.
2. Process and vectorize the content.
3. Upload vectors in batches.
4. Validate successful upload.
5. Provide detailed process logs.

📚 Documentation

Available Tools

The server provides the following tools:

1. search-vectors - Search for similar documents using a text query. Parameters include:

   • query: string (search query text)
   • topK: number (1 - 100, default is 5)
   • filter: object (optional filtering conditions)

2. add-vector - Add a single document. Parameters include:

   • text: string (content for vectorization)
   • metadata: object (vector metadata)
   • id: string (optional custom ID)

3. process-confluence - Process Confluence JSON data. Parameters include:

   • filePath: string (file path)
   • namespace: string (optional, default is "capella-document-search")

4. delete-vectors - Delete vectors. Parameters include:

   • ids: array of strings (list of vector IDs)
   • namespace: string (optional, default is "capella-document-search")

5. get-stats - Get database statistics (currently unavailable).

Database Configuration

The server requires a Pinecone vector database. Configure the connection parameters in the .env file:

PINECONE_API_KEY=Your API key
PINECONE_HOST=Your host address
PINECONE_INDEX_NAME=Your index name
DEFAULT_NAMESPACE=Default namespace

Metadata Schema

Confluence Documents

ID: confluence-[page-id]-[item-id]
title: [Title]
pageId: [Page ID]
spaceKey: [Space key]
type: [Type]
content: [Text content]
author: [Author's name]
source: [Source]