Challenge 1: The Fully Automatic Mode and Why the Companion Doc
Matters Most There -- Solution Walkthrough

Picking the fully automatic "generate from video content" mode with
no script and no live narration means the video is fully exposed to
Chapter 2's structural limitation on Criterion 3. In this mode, there
is no script and no recorded narration for the AI to draw on at all --
its only input is the recorded screen actions and captions, which can
describe what happened (a click, a screen change) but were never
given any information about why a particular approach was chosen.
No amount of AI quality improvement changes this, because the
reasoning was never captured anywhere the pipeline could read it.

This is exactly why the companion document becomes especially
valuable in this specific case: since the video's own narration has no
real intent behind it, the auto-generated written document is the one
place where that missing "why" can still be added after the fact. A
human editor can open the companion doc -- generated from the same
recording, with editable screenshots -- and add the reasoning the
automatic voiceover never had access to, in a format built for direct
text editing rather than requiring a full re-recording or re-narration
of the video itself. The video and the doc end up covering different
gaps: the video shows the actions quickly, and the doc becomes the
place where genuine explanation can be added cheaply.

WHY THIS WORKS AS AN ANSWER
------------------------------
This exercise checks that the reader connects the specific choice of
narration mode to its concrete Criterion 3 consequence, and can
explain precisely why an editable companion document is the natural
fix for exactly this scenario -- not a generic "extra feature," but
the specific answer to the specific gap this mode creates.