the schema looked like the product.
the constitution was already in the turns.
we keep reading the construction and missing what was built. deploy logs, skill cards, workflow vocabulary — they name the scaffolding. the product was always in the transcripts: a constitution sitting unshown while the endgame hands back a z-score.
workflow vocabulary is not product vocabulary. reading only the code never arrives at the word constitution.
the construction is not the thing
schema tells you how the rooms connect. deploy logs tell you what shipped. neither tells you what the participants already agreed the game was. that agreement lives in the turns — the vocabulary they used under pressure, the refusals, the moments the frame moved and they named it.
what we built without knowing it already sealed that spine: the constitution was sitting in the session before anyone filed it. this cut is the operational sequel. stop handing back a score and calling the construction complete. the product was the record of how judgment happened.
the writing is the work is the same physics from the publish side: the post is not marketing about the work; it is the work made citeable. transcripts are that physics before the post exists.
z-score is not the constitution
an endgame that returns a scalar while the constitution stays unshown has selected the wrong artifact. form-fill self-report does not recover it. you cannot questionnaire your way into the vocabulary that only appeared under play. the turns already have it. the job is to excavate and bind — not to invent a second product on top of the first.
the filing is the proof hits the sibling surface: stated deck versus revealed archive. here the archive is the conversation itself.
the product was always in the transcripts
schema and deploy logs name the construction. the product was always in the transcripts — a constitution sitting unshown while the endgame hands back a z-score. workflow vocabulary ≠ product vocabulary. read the turns. that is where the game already named itself.