What it heard
Its reading of your last message, written before it chose a question.
What it picked up
The exact words from your last message that its question is built on. If it can't name any, it wasn't listening.
Paths it considered
Each turn it works out two or three different questions it could ask, then picks one. Both show here.
Still needed
Reading right now
Move—
Engagement—
DepthL1
Question typeFACT
Why it asked that
Its reasoning appears here.
States covered
Emotional range
Feelings get filed under before, during or after — positive and negative treated the same.
Story bank
Usable moments get built out as event, context, emotion, consequence, learning.
The review
Nothing written yet.
What the student has told us
Held back for now
- Questions it could ask but is saving.
Assumption guard clean
Questions that assume a nationality, a home, or that leaving needs justifying get caught here and rewritten before you see them.
Quality signals
Normal
One thread only, and no verification, incentive or consent steps — those are a separate phase and shouldn't contaminate the experience questions.