Notioc MCP Server

The Notioc MCP (Mediated Context Provider) Server is the core backend engine that powers Notioc's ability to understand and interact with your academic context, primarily your Canvas course materials.

Purpose

While Notioc offers a standalone web application, the MCP Server is designed to expose these powerful context-providing capabilities via a secure API. This allows developers and users to integrate Notioc's knowledge directly into other applications, especially Large Language Model (LLM) environments like OpenAI's custom GPTs or other AI assistants.

Instead of just using a generic LLM, you can use the Notioc MCP API to provide the LLM with specific, relevant context from your actual course files, assignments, and potentially discussions, leading to more accurate and helpful responses for your academic tasks.

Key Features (via API)

  • Securely connect to and authenticate with Canvas.
  • List courses, modules, files, and other relevant Canvas resources.
  • Fetch content from various file types (PDF, DOCX, TXT, etc.).
  • Parse and chunk content suitable for RAG (Retrieval-Augmented Generation).
  • Perform semantic searches across your fetched documents.
  • (Potentially) Summarize or perform other basic analysis on fetched content.
  • (Potentially) Perform targeted web searches to augment context.

How to Use

To use the MCP Server API, you will typically need to:

  1. Connect your Canvas account via the Notioc application (details TBD).
  2. Obtain an API key for authentication (see the Authentication section).
  3. Make requests to the documented endpoints (see the API Reference section).

The Notioc MCP API is currently under development. This documentation provides a conceptual overview.