krumjahn/applehealth/apple-health-export-analyzer

Apple Health Export Analyzer

This skill gives OpenClaw two ways to read Apple Health data:

• preferred: use the Health Data AI Analyzer Mac app if its local API is running
• fallback: use the open-source applehealth repo with an export.xml

For Mac app mode in restricted OpenClaw sessions, install the companion plugin from the krumjahn/applehealth repo once:

git clone https://github.com/krumjahn/applehealth.git
cd applehealth
python skills/apple-health-export-analyzer/scripts/install_mac_app_companion.py --restart

That installs:

• the health-analyzer-local OpenClaw plugin
• the health-analyzer-mac-local companion skill for deterministic localhost access

Use it when the user wants one of these outcomes:

• verify the local setup
• generate a daily brief from recent Apple Health data
• compare steps and sleep over the last 7 days
• produce a weekly summary with workouts and heart rate context
• set up a recurring daily health message with OpenClaw heartbeat or automation

Quick workflow

1. First try --source auto.
2. If the Mac app local API is reachable, use that.
3. Otherwise use the applehealth repo plus export.xml.
4. Verify setup with scripts/check_setup.py.
5. For a daily brief, use scripts/daily_brief.py.
6. For a 7-day comparison, use scripts/compare_recent_trends.py.
7. For a weekly summary, use scripts/weekly_summary.py.

If the repo is not in the current workspace, ask the user to clone:

git clone https://github.com/krumjahn/applehealth.git

If needed, read:

• workflow for exact commands and expected outputs
• troubleshooting when setup fails
• heartbeat for recurring daily messages

Sources

Option 1: Health Data AI Analyzer Mac app

• local API running on the Mac
• app has a saved integration dataset
• no export.xml needed
• for TUI sessions without local file/shell tools, clone https://github.com/krumjahn/applehealth, install the companion plugin, and then use health-analyzer-mac-local
• the public skill uses read-only /openclaw/* localhost endpoints for this mode and does not read local token files

Option 2: applehealth repo

• a cloned applehealth repo
• an Apple Health export.xml

Optional:

• APPLEHEALTH_REPO environment variable pointing to the repo root
• --repo, --export, and --out flags for the bundled scripts
• --source auto|mac-app|export

Core commands

Verify setup

python skills/apple-health-export-analyzer/scripts/check_setup.py --source auto --repo /path/to/applehealth --export /path/to/export.xml --out /path/to/analysis

Daily brief

python skills/apple-health-export-analyzer/scripts/daily_brief.py --source auto --repo /path/to/applehealth --export /path/to/export.xml --out /path/to/analysis --date 2026-03-19

Compare recent trends

python skills/apple-health-export-analyzer/scripts/compare_recent_trends.py --source auto --repo /path/to/applehealth --export /path/to/export.xml --out /path/to/analysis --days 7

Weekly summary

python skills/apple-health-export-analyzer/scripts/weekly_summary.py --source auto --repo /path/to/applehealth --export /path/to/export.xml --out /path/to/analysis --days 7

Response style

When the scripts return JSON:

• summarize it into a concise health brief
• call out missing data explicitly
• give practical suggestions, not medical advice
• do not invent metrics that are absent

For daily brief responses, prefer this structure:

• Status
• What changed
• 3 suggestions
• Missing data

Example output

Status
- Activity was below your recent baseline.
 
What changed
- Steps: 2,444 vs 7-day baseline 10,005
- Workouts: 0 minutes, 0 workouts
- Sleep: insufficient data
 
Suggestions
1. Add one easy walk instead of trying to catch up with a hard workout.
2. Use one fixed movement anchor like a walk after lunch.
3. Keep effort moderate when recovery data is missing.