pinecone-vector-db-mcp-server
If you are the rightful owner of pinecone-vector-db-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.
This project implements a Model Context Protocol (MCP) server for interacting with a Pinecone vector database, supporting RAG-processed PDF and Confluence data.
The MCP Pinecone Vector Database Server is designed to facilitate the reading and writing of vectorized information to a Pinecone vector database. It supports operations such as searching for similar documents using text queries, adding new vectors with custom metadata, processing and uploading Confluence data in batches, and deleting vectors by ID. The server is built to handle RAG-processed PDF data and Confluence data, making it versatile for various document management needs. The server requires a Pinecone API key and an OpenAI API key for generating embeddings, and it operates using the Bun runtime. Users can configure the server through a `.env` file, specifying connection details for the Pinecone database.
Features
- Search for similar documents using text queries
- Add new vectors to the database with custom metadata
- Process and upload Confluence data in batch
- Delete vectors by ID
- Basic database statistics (temporarily disabled)
Tools
search-vectors
Search for similar documents with parameters: query, topK, and filter.
add-vector
Add a single document with parameters: text, metadata, and id.
process-confluence
Process Confluence JSON data with parameters: filePath and namespace.
delete-vectors
Delete vectors with parameters: ids and namespace.
get-stats
Get database statistics (temporarily disabled).
