Tools · Toddler years
Tantrum Coach
Talk you through an active meltdown, in the moment.
How it works
Grounded generation. Describe a toddler tantrum as it's happening; the service returns a staged, connection-before-correction de-escalation script — regulate your own body and voice first, name the feeling out loud, hold the limit while staying warm, then redirect and repair — with concrete lines to say for THIS scenario, what to avoid because it pours fuel on it, and what to do once they're calm. Grounded in the emotion-coaching corpus (Gottman, Siegel/Bryson, Faber/King), not generic advice. For The Family Almanac's toddler guide.
You bring
{ scenario, child_age_years?, trigger?, cluster?, model? }
You get
{ scenario, stages[] (step·do·say·why), avoid[], after, principle, grounded_in, provenance }
Use it for
- →Screaming on the floor because we left the park → four-stage script + escalators to avoid + a repair step
- →Bedtime meltdown, hitting and refusing pajamas → co-regulation-first script that holds the limit while validating the upset
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.