Morning flow
Generated per user at open time. Each emotion family keeps its own active master prompt; the reader's picked emotion selects the family via the intro config's flow tags, and Default is the fallback.
Master prompt
The prompt document the generator runs. Versioned, with one active per emotion family — the reader's picked emotion selects the family; Default is the fallback.
Request framing
What wraps the master prompt on every morning generation. Edits apply to the NEXT generation immediately — nothing is cached.
Runtime reminder — appended to the user message
The format + budget instruction the model
reads last. Keep {{screenBudget}} where this
run's screen ceiling should go — it is substituted at
generation time with the live number below.
Assembled user message preview (read-only)
Exactly what the model receives as the user message. The [bracketed] parts are dynamic per generation; the computed budget is real.
Presentation appendix — composed onto the system message
The VOICE COMPRESSION + OUTPUT FORMAT (screens and blocks) spec. It is joined after the CEO method body of the active master prompt. A self-contained prompt (one that already carries the OUTPUT FORMAT inline) is used as-is and this appendix is NOT appended until a method-only prompt is active.
Composed system message preview (read-only)
The active method body and this appendix, labeled. Edit the method in the Master prompt editor above; edit the appendix here.
Intro screens
Describe the intro screens in plain language and
the AI drafts a valid config. Preview the tap-through, fine-tune
the JSON if needed, save as a draft, then activate. Versioned;
exactly one active. Emotion options may carry a
"flow": "memorial" | "dislocation" tag that picks
which master-prompt family generates the flow.
Live intro preview
Tap through the preset input kit: pick an emotion, fill the conditional sentence, then see the captured variables map and the assembled prompt input.
Model + cost
Pick the brain the test generation below runs on, see what a run should cost before you spend it, and correct the price table without a deploy.
The selector alone changes nothing on the server — it rides along with the next test generation as a one-off override. Only the save button above changes what real users get.
Projected cost — estimate, before you run
Run history — this session only
Price table
USD per 1,000,000 tokens. Leave a cell blank when a provider does not bill that dimension (OpenAI has no separate cache-write charge). Saving marks the table verified — that badge is what every cost number on this page is qualified by.
Generate + simulate a devotional
Tap the intro preview above through to its review panel, then generate the real devotional from those captured answers. This runs the live master prompt (one AI call — it can take a while) and deterministically wraps the result into the tap-through flow. Nothing is saved.
A testing aid: paste prior days' ledger lines to exercise day-2+ behavior — the day number should increment and the mirror should differ (every prior mirror is excluded).
Generated devotional
Tap through the real flow — including the card-30 figure prompt feeding the card-31 prayer (the one orange token replay).
What this run cost
Ledger line — machine memory, never shown to the user
Warnings
Raw generated script (pre-wrap)
Evening flows
The daily evening flow - same for everyone, delivered exactly as published per date.