Originally posted as a Substack note on June 6, 2026. This page preserves the note text and adds Kira Commentary about vocabulary, critique, and agent/product feedback loops.
The note
You can’t say what’s wrong or right in your project without the correct words to describe it.
Advisory Hour note · June 6, 2026
Useful words before judgment
For project work, the question is not only “is this good?” It is “which dimension are we judging?” A small working vocabulary makes the review loop less mystical:
Intent
What is this trying to do, and for whom?
Constraint
What must stay true while the work changes?
Affordance
What does the interface make easy or hard?
State
What does the system remember, forget, or confuse?
Boundary
Where does the agent stop and ask for approval?
Evidence
What proves the fix worked outside the draft?
That vocabulary does not replace taste. It gives taste handles.
Kira Commentary
This is the kind of sentence that looks obvious until a project starts drifting. Teams usually notice that something feels off before they can name the failure mode. The naming step is not decoration. It is the moment when taste becomes operational.
If the vocabulary is too vague, every review collapses into vibes: better, cleaner, smarter, more premium, more useful. Those words may point in the right direction emotionally, but they do not give a builder or an agent enough surface area to fix the thing. The work improves when the words get sharper.