What the Agent Assumed
Every silent decision the agent made, pinned and counted.
every silent decision the agent made is pinned directly onto the interface element it touched — the invisible gap-filling AI prototyping tools do by default becomes an annotated list instead of a surprise found in QA
What goes in, what comes out
Learner writes a deliberately incomplete prototyping brief, as real briefs are, and sends it to a live coding/design agent that builds an interactive prototype. Every place the agent had to invent a decision the brief never specified gets logged as an assumption event pinned to the exact element it affected.
write the brief; send to the agent; review the built prototype; accept or reject each logged assumption
the working prototype plus an assumption-log sidebar (e.g. 'you didn't specify an empty state, the agent invented one') scored against the learner's real intent
brief-completeness pre-check meter before send; component complexity (single screen / 3-screen flow)
Module 'Co-Building, Not Just Prompting' L2 demo-to-lab pair
Go deeper, elsewhere
Hand-picked public explainers and open tools that complement this one — always optional, never required, never graded.
Concepts click when you open the machinery.
Three labs are already live and free — the same hands-on style this playground brings to its module.