Blickwechsel banner
CUHK-Business-School-AI-Hub CUHK-Business-School-AI-Hub

Blickwechsel

Development community

Description

A change of perspective on your own teaching. Audits a lecture you already teach from the student's point of view, then rebuilds what you accept: activities, rotatable datasets, productive-failure cases, and edits to the slide file itself. One Claude skill.

Installation

This entry records only its repository, not the path inside it, so there is no exact command to give. Open the source below and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

README

BLICKWECHSEL

*A change of perspective — your lecture seen from where the student sits. The best thing in it is usually not in the file.*


It does not rewrite your lecture

Ask an AI to improve a lecture and you get something back that is tidy, confident, plausible, and quietly worse. Not because it invented a fact. Because it smoothed out the demonstration you always do on the tablet, merged the slide you spend ten minutes on with the one you skip, and deleted the story you tell every year — none of which were written down anywhere it could see.

This skill is built around that failure. It looks first, says what it measured and what it guessed, asks you about the parts it cannot see, and only then proposes anything. You keep every decision that matters.

Everything it says is labelled **measured**, **inferred**, or **assumed**, so you can tell at a glance which of its claims survive if it misunderstood your subject.

It looks before it suggests

The first thing it does is count. It opens PowerPoint, PDF, LaTeX, notebooks or a whole course folder and reports what is actually there: how many slides repeat the one before, which questions are asked with no way to answer them, which slides carry more words than anyone can read and listen to at once, and how much of a notebook the student actually writes rather than watches.

Then it reads the whole thing again as somebody who does not yet know the answer. Where would I first get lost? What am I assumed to already know? What can I do at the end that I could not do at the start?

Only after that does it propose anything — numbered, so you can accept or reject each one, with a reason recorded for every rejection.

It assumes your students already have an AI

Not that they might. That they do.

So a rule telling them not to use one is never treated as a control. Instead it sorts your learning objectives by a blunt question — *could a student produce a passing answer to this in under five minutes with a