knowledge-mcp-server

knowledge-mcp-server

3.2

If you are the rightful owner of knowledge-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 for managing and querying knowledge bases through vector databases, built on top of the dynamic-mcp-server (DMS) framework.

The Knowledge MCP Server is designed to manage and query knowledge bases using vector databases, leveraging the dynamic-mcp-server (DMS) framework. It provides a secure, user-aware, and dynamic platform for AI tool servers, handling user, tool, and database management. The server integrates with MongoDB for centralized database connections and supports extensibility through subclassing and DMS APIs. It offers features like vector database integration, document processing, and secure access control. The server supports Microsoft OneDrive as a knowledge source and plans to include website crawling in future updates. It allows for dynamic tool registration and management, ensuring secure access and sharing of knowledge sources.

Features

  • Vector database integration for storing and querying embeddings
  • Document processing pipeline for content ingestion
  • Dynamic tool registration and management via DMS APIs
  • Secure access control and sharing for knowledge sources
  • MongoDB integration for metadata and vector storage

Tools

  1. Add Knowledge

    Ingests and processes new documents into a knowledge source.

  2. Search

    Queries a specific knowledge source for relevant information.

  3. Use Knowledge Source

    Creates a new tool for interacting with a specific knowledge source.

  4. Refresh Knowledge Source

    Refreshes a knowledge source by re-ingesting its content.