Educator desk run, not a classroom trial. I followed Day 91 with A printed: set up takes the drawing plus NOT TOLD card; its rectangle/circle/waves are large enough to point to. I answered “who made it?” from A alone and correctly had to stop. Day 93 separates Ivo's reason from all-visitor outcome. Day 94 makes a copied clause physically traceable to B. Day 96 introduces Tala as one limited account rather than “the sign failed.” Day 99 permits a private adult listener instead of a performance. The two checks use unseen names and objects, and their worked answers remain in teacher/. A teacher still needs to fit these optional blocks to actual allocation and assess authentic K01–K04 elsewhere.
Learner desk simulation, not a child result. I tried the printed, oral and no-print routes. For Day 91 I felt a rectangle, one round token and three strings, then placed a question card face down for an unknown maker. For Day 97 I matched “what is drawn?” to A and “why did Ivo say?” to B, and left “everyone?” unfilled. For Check B I could correctly say “Ellis asked Una; this is one person” without seeing the key. The route does not ask a child to disclose home or family details. Actual child comprehension, accessibility and engagement are untested.
Family/carer desk pass. Every bridge is 2–5 minutes and optional; no printing, phone, money, translation, object brought from home, answer return or adult availability is assumed. Skipping cannot block the next class lesson. Any home-language version requires a competent reviewer before publication; none is supplied here.
Potential funder desk pass. The package contains ten timed days, 30 route choices, 20 worked transfers, 10 home options, eight original illustrated A4 pairs with exact alternatives, two separate public checks/key, official source pins, rights ledger and content-hash manifest. The visible output supports a claim of authored draft completeness and reproducibility. It does not support a claim that children learned, teachers saved time, accessibility worked for every learner, a community approved local content, or that a school year is complete. Those require actual implementation, consented evidence and local review.
Self-critique and next revision: The deliberate fiction makes the source boundary clear but supplies no authentic knowledge evidence; a school must use the local gate. Eight print pages add a preparation burden; offer the exact one-page linear/no-print reading and shape tokens. Teacher-only key separation reduces accidental answer modelling but public checks are not secure. Revisit wording and contrast with actual users before claiming 10/10 experience.