Python Claude ChatGPT HEMIS AsyncIO REST APIs MCP

View on GitHub:
github.com/sukhrobyangibaev/mcp_hemis_student

HEMIS Student MCP is an open-source Model Context Protocol (MCP) server that lets you securely connect your university’s HEMIS student portal to Claude or ChatGPT on your desktop. Instantly access your grades, schedule, attendance, contracts, and official documents—right from your favorite AI assistant, with full privacy and no third-party sharing.

Built in Python, this project leverages async REST APIs and the MCP standard to bridge your university’s HEMIS system with modern AI chat clients. It’s perfect for students who want instant, private access to their academic data, or for universities looking to offer AI-powered self-service.

Key Features

  • Personal Academic Dashboard:
    View your student profile, GPA, grades, and semester history in natural language.
  • Class Schedules & Attendance:
    Instantly check your weekly schedule, attendance records, and exam timetables.
  • Official Documents & Contracts:
    Download diplomas, transcripts, references, and contract info directly from HEMIS.
  • Task & Resource Management:
    See assignments, deadlines, electronic resources, and performance stats for each subject.
  • University Statistics:
    Explore public stats about students, employees, and university structure.
  • Secure & Private:
    Your HEMIS credentials are stored locally and never leave your machine.

Technical Architecture

  • Backend:
    Written in Python 3.10+, using httpx for async REST calls and mcp for the Model Context Protocol server.
  • Integration:
    Works with both Claude for Desktop and ChatGPT Desktop via MCP, exposing all tools as callable functions.
  • Authentication:
    Uses environment variables for HEMIS API base, login, and password. Tokens are cached locally for efficiency.
  • Extensible:
    Easily add new endpoints or customize tool outputs for your university’s needs.
  • Client Examples:
    Includes sample clients for both OpenAI and Anthropic APIs.

Example Workflow

  1. Configure Claude or ChatGPT Desktop:
    Add the MCP server to your desktop AI config, providing your HEMIS credentials.
  2. Start the MCP Server:
    Run the server with uv run server.py (or via the desktop app).
  3. Ask Your AI Assistant:
    • “Show me my current GPA”
    • “What classes do I have this week?”
    • “Download my transcript”
    • “Show my attendance for Databases”
  4. Get Instant Results:
    The AI calls the MCP tools, fetches your data from HEMIS, and presents it in chat.

Technologies Used

  • Python 3.10+
  • httpx
  • mcp (Model Context Protocol)
  • dotenv
  • AsyncIO
  • REST APIs (HEMIS)
  • OpenAI / Anthropic SDKs

This project shows how open protocols and async Python can empower students to take control of their academic data—securely, privately, and with the help of modern AI.

Try it out or contribute:
github.com/sukhrobyangibaev/mcp_hemis_student