tue-api-wrapper, created by SebastianBoehler, is an MCP server that connects LLM-based assistants to University of Tübingen academic systems to automate study tasks. The wrapper exposes campus services as callable tools so models can fetch schedules, course materials, lecture recordings, and grade metadata. It ships with a Python SDK and supports MCP-compatible clients, targeting students and staff who want conversational access to university data and scripted automation of study workflows.

What tasks can you actually use it for?

The wrapper turns campus systems into tool endpoints so language models can perform concrete study tasks. Examples include summarizing course schedules, checking for new lecture recordings, and retrieving course documents. Supported systems listed by the project include Alma for registration and grades, ILIAS and Moodle for learning materials, and TIMMS for recordings. Use cases are automation of routine queries and enabling conversational agents to access campus resources.

How reliable are the data accesses and responses?

The wrapper’s responses reflect the source systems it queries rather than model inference alone; reliability depends on the underlying APIs and their current data. The project is noted within the MCP community as a practical demonstration of protocol use, which indicates functional interoperability. Users should treat returned information as a data retrieval result that may require verification against the original system before acting on sensitive academic decisions.

What inputs and setup does it require?

Deployment requires a Python runtime and an MCP-capable client such as Claude Desktop or a custom agent. Installation paths include pip from the repository’s package subdirectory. The server runs as an MCP endpoint and needs configuration of university credentials, typically supplied via environment variables, and network access to the campus APIs you intend to query.

How does it fit into student workflows and institutional context?

The project consolidates multiple fragmented campus systems into a single AI-accessible interface as part of the Tue-StudyOS initiative, which can reduce manual lookups across platforms. It suits workflows that accept a hosted service layer between agents and campus APIs, and it can be used as a standalone Python library or as a long-running MCP server integrated into existing agent setups.

Best for technically minded Tübingen users who accept self-hosting responsibility

The wrapper is a practical option for students and staff who want programmatic, conversational access to campus data and who can manage a Python service and MCP client. Automated security audits of the project highlight the need for careful credential practices, so adopt least-privilege accounts and segregated service credentials. For those prepared to self-host and verify returned data, the wrapper delivers targeted integration value for academic workflows.