Living Word Lesson Plans — Data-Driven Per-Scene + Printable Export (Phase D-5)
STATUS: DRAFT. The deep-dive plan's "artifact that converts": a teacher can't currently answer "what do I do for the next 30-minute session?" — so the game reads as a supplement, and supplements don't get curriculum budget. This makes each scene runnable as a lesson. Curriculum content = founder review gate (theological/pedagogical), like the lens curriculum.
§1 — Purpose + the adoption gap
Promote the Eden-only "Classroom Run" (data/classroom.ts — verify its current shape before building; do not assume) to a data-driven lesson-plans.ts keyed by scene, each a ready-to-run 30-minute session: objective, Scripture passage, time-boxed beats (open → play → discuss → close), discussion prompts, and a printable export. Flip the value from "author your own quizzes" to "faithful, ready-to-run, denomination-fit curriculum you can extend."
§2 — What exists / what's needed
- Have: scenes (SCENE_REGISTRY), per-tradition lens questions (shipped), reviewed quiz content.
- Need: a typed
LessonPlanper scene + the authored CONTENT (the real work, founder-reviewed) + a teacher-facing render + a print/PDF-friendly export.
§3 — Decisions (founder)
| # | Decision | Notes |
|---|---|---|
| LP1 | Lesson-plan structure | Proposed: { sceneId, title, objective, scriptureRef, durationMin, beats: {label, minutes, teacherNote}[], discussionPrompts[], closingPrayerOrReflection }. Confirm/adjust fields. |
| LP2 | Which scenes get plans first | All 16, or start with the flagship arc (Eden/Flood/Babel/Abraham/Exodus)? |
| LP3 | Lens-awareness | Should a plan adapt to the classroom's tradition_lens_id (e.g. surface that tradition's lens question in discussion)? Ties to the shipped lens curriculum. |
| LP4 | Age-banding | One plan per scene, or per age band (Preschool/Elementary/Preteen/Youth/Adult)? The plan flags named age tracks as the right move. |
| LP5 | Export format | Printable HTML print-view (cheap, recommended v1) vs generated PDF. |
§4 — Content authoring (the real cost)
The schema is small; the content is the work and must be founder-authored/reviewed (faithful, tradition-fair, AI-Bridge-aligned — the game/material points to the living teacher, never poses as the doctrinal arbiter). Recommend authoring with the same discipline as the lens curriculum (founder-review gate per scene). Could draft candidate plans via claude -p (CLAUDE.md rule #5 — CLI, never API/chatbot endpoint) for founder review, never auto-ship.
§5 — Recommendation
Lock LP1 (structure) + LP2 (start with the flagship arc) + LP5 (print-view) first; build the typed model + render + print-view against ONE founder-authored sample plan (Eden), verify, then scale content scene-by-scene through the review gate. Pairs with the scope-and-sequence spec.