healthkit-cli（healthkit CLI 命令行）

Use bunx apple-health to interact with HealthKit through a running Expo app's devtools connection.

Prerequisites

The app must have the devtools hook enabled:

import { useHealthKitDevTools } from "apple-health/dev-tools";

export default function App() {
  useHealthKitDevTools();
  // ...
}

Verify connection:

bunx apple-health status

Seeding Data

Quick Start

Use batch mode for efficient bulk writes. Create NDJSON data and pipe to the CLI:

cat << 'EOF' | bunx apple-health batch
{"kind":"quantity","type":"heartRate","value":72,"start":"today 8am"}
{"kind":"quantity","type":"stepCount","value":8500,"start":"yesterday","duration":"1d"}
{"kind":"category","type":"sleepAnalysis","value":4,"start":"-8h","duration":"2h"}
{"kind":"workout","activityType":"running","start":"-2h","duration":"45m","energy":450}
EOF

Sample Types

Quantity samples (measurements with values):

{"kind":"quantity","type":"heartRate","value":72,"start":"today 8am"}
{"kind":"quantity","type":"stepCount","value":10000,"start":"yesterday","duration":"1d"}
{"kind":"quantity","type":"dietaryCaffeine","value":150,"start":"today 7am"}
{"kind":"quantity","type":"activeEnergyBurned","value":350,"start":"today","duration":"1d"}
{"kind":"quantity","type":"bodyMass","value":75,"unit":"kg","start":"today 7am"}

Category samples (events/states with enum values):

{"kind":"category","type":"sleepAnalysis","value":3,"start":"-7h","duration":"6h"}
{"kind":"category","type":"headache","value":2,"start":"today 2pm"}
{"kind":"category","type":"mindfulSession","value":0,"start":"-1h","duration":"15m"}

Workouts:

{"kind":"workout","activityType":"running","start":"-1h","duration":"30m","energy":300,"distance":5000}
{"kind":"workout","activityType":"cycling","start":"today 7am","duration":"1h","energy":500,"distance":25000}
{"kind":"workout","activityType":"yoga","start":"yesterday 6am","duration":"45m","energy":150}

Sleep Values

Realistic sleep pattern example:

{"kind":"category","type":"sleepAnalysis","value":0,"start":"-8h","duration":"8h"}
{"kind":"category","type":"sleepAnalysis","value":3,"start":"-7h45m","duration":"45m"}
{"kind":"category","type":"sleepAnalysis","value":4,"start":"-7h","duration":"1h"}
{"kind":"category","type":"sleepAnalysis","value":5,"start":"-6h","duration":"30m"}
{"kind":"category","type":"sleepAnalysis","value":3,"start":"-5h30m","duration":"2h"}
{"kind":"category","type":"sleepAnalysis","value":5,"start":"-3h30m","duration":"45m"}
{"kind":"category","type":"sleepAnalysis","value":3,"start":"-2h45m","duration":"2h"}

Symptom Values

Date Formats

Duration uses same format: "duration":"1h30m", "duration":"1d", etc.

Data Profiles

Reference values for realistic data generation:

Healthy Active Person

• Resting HR: 55-65 bpm
• Steps: 8,000-12,000/day
• Sleep: 7-8 hours, good quality
• Workouts: 4-5x/week
• Water: 2-3L/day

Sedentary Office Worker

• Resting HR: 70-80 bpm
• Steps: 2,000-4,000/day
• Sleep: 5-6 hours, fragmented
• Workouts: 0-1x/week
• Caffeine: 300-500mg/day

Elite Athlete

• Resting HR: 45-55 bpm
• Steps: 15,000-25,000/day
• Sleep: 8-9 hours, high quality
• Workouts: 10-14x/week (doubles)

Stressed Individual

• Resting HR: 80-95 bpm
• Steps: 2,000-3,500/day
• Sleep: 4-5 hours, poor quality
• Symptoms: headaches, fatigue
• Caffeine: 400-600mg/day

Verifying Data

Query Samples

# Query recent samples
bunx apple-health query quantity heartRate --limit 10
bunx apple-health query category sleepAnalysis --limit 5
bunx apple-health query workouts --limit 5

# With date range
bunx apple-health query quantity stepCount --start "-7d" --end "now" --limit 100

Get Statistics

# Single stat
bunx apple-health stats stepCount

# With aggregations
bunx apple-health stats heartRate --aggregations "discreteAverage,discreteMin,discreteMax"

# Time-bucketed (daily, weekly, etc.)
bunx apple-health stats stepCount --interval day --start "-7d"
bunx apple-health stats heartRate --interval hour --start "today"

JSON Output

Add --json flag for machine-readable output:

bunx apple-health query quantity heartRate --limit 5 --json
bunx apple-health stats stepCount --interval day --start "-7d" --json

Individual Writes

For single samples without batch mode:

# Quantity samples
bunx apple-health write quantity heartRate 72
bunx apple-health write quantity heartRate 85 --start "today 8am"
bunx apple-health write quantity stepCount 5000 --start "yesterday" --duration "1d"

# Category samples
bunx apple-health write category sleepAnalysis 3 --start "-8h" --duration "7h"
bunx apple-health write category headache 2 --start "-2h"

# Workouts
bunx apple-health write workout running
bunx apple-health write workout cycling --start "today 7am" --duration "1h" --energy 450 --distance 25000

Deleting Test Data

# Delete specific type within time range
bunx apple-health delete stepCount --start "-30d" --end "now"
bunx apple-health delete heartRate --start "-30d" --end "now"

Authorization

Check and request permissions before writing:

# Check authorization status
bunx apple-health auth status stepCount heartRate

# Request authorization
bunx apple-health auth request --read "stepCount,heartRate" --write "stepCount,heartRate"

Available Types

List all available types:

bunx apple-health types

# Filter by category
bunx apple-health types --category Vitals
bunx apple-health types --category Nutrition

Common quantity types:

• Body: bodyMass, height, bodyFatPercentage, bodyMassIndex
• Fitness: stepCount, distanceWalkingRunning, activeEnergyBurned, flightsClimbed
• Vitals: heartRate, restingHeartRate, bloodPressureSystolic, oxygenSaturation
• Nutrition: dietaryCaffeine, dietaryWater, dietaryEnergyConsumed, dietaryProtein

Common category types:

• Sleep: sleepAnalysis
• Symptoms: headache, fatigue, nausea, dizziness
• Mindfulness: mindfulSession

Workout types: running, walking, cycling, swimming, yoga, hiking, highIntensityIntervalTraining, traditionalStrengthTraining, and 70+ more.

Tips for Realistic Data

1. Include variations: Not every day should have identical values
2. Consider correlations: Poor sleep → higher resting HR, lower step count
3. Use appropriate ranges: Elite athletes have lower resting HR than sedentary individuals
4. Add realistic patterns: Heart rate higher during workouts, lower during sleep
5. Distribute over time: Use relative time formats to spread data across days

Interactive Mode

For exploratory testing, use the REPL:

bunx apple-health repl

apple-health> write quantity heartRate 72
apple-health> query quantity heartRate 5
apple-health> stats stepCount day
apple-health> exit