Tools · General business
Sleep Schedule Builder
Build today's baby nap/bedtime schedule from age + this morning's wake time.
How it works
Deterministic wake-window scheduling (compute.ts) OVER a corpus-grounded note. Age-appropriate wake windows + nap counts (the widely-cited Healthy Sleep Habits / Precious Little Sleep ranges) are encoded as code; from the morning wake time it lays out naps and bedtime with overtired warnings, and the LLM only writes a short how-to-use note (catch the pre-overtired window, read the baby) grounded in the newborn sleep corpus. For The Family Almanac's newborn guide.
You bring
{ age_months, wake_time ('HH:MM' 24h), cluster?, model? }
You get
{ age_band, wake_window, recommended_naps, schedule[], overtired_warnings[], guidance, grounded_in, provenance }
Use it for
- →6-month-old woke at 7:00 → nap/bedtime plan with overtired-loop warnings
- →Give an exhausted parent a concrete starting frame instead of guessing wake windows
Run it on your data
Call it on your own inputs — over the API, or hand it to your AI agent via MCP. Discovery is open; running it is metered.