# Desk run-through · teacher, learner and potential funder views

**Scope of this review:** desk reading and deterministic artifact inspection only, 30 September 2026. No child, teacher, family, local First Nations authority, independent accessibility specialist or funder has piloted this pack. No candidate audio has been approved by human listening. The checks are diagnostic practice, not formal instruments.

## Teacher with 25 minutes

Day 13: collect C print or four broad X cards, two gap strips and the optional audio index. A teacher can say “Both rows have four; this plan has more time between beats,” then run 3/4/5/7/4/2 minutes. If speakers are unavailable or the candidate cues are not yet human-approved, pointing at two different regular paces and labelled gap strips completes the *planning* activity. The teacher records count/tempo reasoning and does not tick “heard slow versus quick.” This is ready for a desk-run lesson but still needs school-level access and sound-source review.

Day 19: the educator can use a child's own score and optional willing adult without copying music. If a suitable real work has not been authorised and contextualised, the “maker and community” encounter stays on hold; the educator cannot claim AC9AMUFE01 from our synthetic cues. Sharing is logged as actually happened: original score shown, sound heard, puppet rehearsal or pass.

## Learner using a quiet route

The Day 16 D page has two aligned four-place rows. A child who does not want sound can direct a neutral adult to place four X above and one X below at count 3, then explain that pulse continues while answer enters once. This demonstrates the score concept, not real-time rhythmic coordination. Wide cells, a column straightedge, accessible text/tactile directions and the option to pass reduce unnecessary sensory and motor demands; classroom suitability is still locally checked.

## Assessor or funder audit

The [crosswalk](CURRICULUM-CROSSWALK.md) quotes exact four QCAA Prep Music content descriptions and labels partial evidence. First responses for two new checks are held separately from keys; I needs a star relocation from 2 to 4 while four pulse X stay, J needs a lower-row answer at 4 after pulse 3. [Source/rights](SOURCE-AND-RIGHTS.md) separates original assets from official documents and a held real-work slot. The read-only verifier checks timings, rows, assets, PDF geometry/text/font, candidate PCM, Markdown list spacing and SHA-256 hashes. Passing it is an artifact quality check, not an impact or accessibility claim.

## Known improvements before classroom roll-out

An educator needs to inspect the printed A4 at actual scale and adapt for specific learners, hear all candidate cues at school playback volume or leave them unused, review safe sound sources and the room's noise level, select an authorised real work with appropriate cultural context, and record actual audience consent. A class pilot with children and teachers, accessible media testing, and language translation review remain open. No neurotype-specific assumptions or compulsory “learning style” labels are made.

## Full-year review repair · 30 September 2026

The independent agent's twenty-pack Music review found a reproducibility defect here: the manifest included three disposable lint-cache entries. The audit now excludes lint, test and Python cache directories from its resource inventory; the intentionally refreshed manifest covers 38 actual resource files. Root ran the complete default audit and a temporary-copy check: different contents in each of those three cache directories leave the resource hashes unchanged, while a change to the learner checks still changes them. No lesson, answer, print page or candidate audio was changed. This repairs publication integrity; it does not clear the remaining classroom or audio gates.