| name | only-baby-skills |
| description | Analyze contraction JSON and baby log JSON to assess mum's labour/contraction situation and baby's feeding and diaper status. Use when the user provides (or references) contractions_*.json and babyLogs_*.json files and wants to know if mum is safe and baby is healthy, or asks for a summary of contractions, feeding, or diaper changes. |
Analyze Mum & Baby Logs
Analyze two JSON data sources to produce a safety and health summary: mum's contraction situation and baby's milk feeding and diaper change status. Always end with a clear verdict and any recommendations.
When to Use
- User provides or references contraction and baby log JSON files (e.g.
contractions_*.json, babyLogs_*.json).
- User asks whether mum is safe, baby is healthy, or for a summary of contractions / feeding / diapers.
Input Files
- Contractions JSON – Array of objects with
id, startTime, endTime (ISO 8601).
- Baby logs JSON – Object with
babyLog (array of entries) and birthday (ISO 8601). Each entry has id, timestamp, type, and type-specific details:
type === "feeding": feedingDetails.volumeML (number, mL).type === "diaper": diaperDetails.hasPee, diaperDetails.hasPoo (booleans).type === "breastFeeding": breastFeedingDetails.durationSeconds (number).
See references/schemas-and-thresholds.md for exact schemas and health/safety thresholds.
Workflow
1. Load and parse both JSON files
- Resolve file paths from user message or workspace (e.g. Downloads, project paths).
- Parse contractions as an array; baby data as object with
babyLog and birthday.
2. Analyze contractions
- Sort contractions by
startTime (ascending = chronological).
- For each contraction compute duration =
endTime - startTime (in seconds/minutes).
- Compute interval = time from previous contraction's
endTime to current startTime (minutes). For the first contraction, interval is N/A.
- Summarize: count, time range of data, typical duration, typical interval, and any pattern (e.g. regular vs irregular).
- Apply safety rules from references/schemas-and-thresholds.md (e.g. 5-1-1 rule, when to seek care).
3. Analyze baby logs
- From
birthday and latest log timestamp, compute baby's age (e.g. days or weeks).
- Split
babyLog by type: feeding, diaper, and breastFeeding.
- Feeding (bottle/expressed): extract
feedingDetails.volumeML and timestamp. Compute total volume and feed count over last 24 h (and optionally last 48 h). Compute average volume per feed, average volume per hour (total mL / hours in window), and approximate interval between feeds.
- BreastFeeding: extract
breastFeedingDetails.durationSeconds and timestamp. Compute session count and total duration (e.g. total minutes) over last 24 h (and optionally last 48 h). Optionally report average session length.
- Diaper: count wet (
diaperDetails.hasPee), dirty (diaperDetails.hasPoo), and both. Compute counts over last 24 h (and optionally last 48 h).
- Compare to age-appropriate thresholds in references/schemas-and-thresholds.md (feeds/sessions per day, wet/dirty diaper expectations).
4. Produce report and verdict
Output:
-
Mum – Contraction summary
Count, date range, duration/interval stats, pattern. Then: Mum safe? (Yes / Monitor / Seek care) with short reason and any next step (e.g. "Continue timing; if 5-1-1, go to hospital").
-
Baby – Feeding & diaper summary
Age; bottle feeds in last 24 h (count + total mL); breastfeeding sessions in last 24 h (count + total duration if present); diapers in last 24 h (wet/dirty). Then: Baby healthy? (Yes / Monitor / Concern) with short reason and any recommendation (e.g. "Ensure 8+ feeds and 6+ wet diapers per day").
-
Caveat
One line: this is not medical advice; when in doubt, contact a midwife, OB, or paediatrician.
Output format
Use clear headings and bullet points. Lead with the two verdicts (mum safe? baby healthy?) then expand with numbers and brief reasoning. Keep the report scannable and under one screen where possible.
