The minimum viable product is the smallest thing that can prove or disprove the riskiest assumption. It is a measuring instrument, not a small version of the final product.
Decide what result would make you stop before you build it. An MVP with no falsifying outcome will always succeed, and teach you nothing.
In practice: Makerz Network
The first thing built was not the platform. It was an assistant for writing user stories — one artefact, and specifically the one that discovery had already measured at around two hours a day.
That is the right size for a measuring instrument. If assisted drafting does not move those two hours, then the claim that artefact production is the bottleneck is wrong — and it is wrong cheaply, before there is a marketplace to be wrong about.