Export everything
A copy of all your health data, every app in one set of files: CSV to open in a spreadsheet, Parquet for pandas, Polars, DuckDB or Spark.
A copy of all your health data, every app in one set of files: CSV to open in a spreadsheet, Parquet for pandas, Polars, DuckDB or Spark. Same columns whatever device the data came from.
python3 recipes/export-all.py # CSV, into health-export-<date>/
python3 recipes/export-all.py --format parquet --since 2026-01-01health-export-2026-10-12/daily.csv 22,907 bytes
health-export-2026-10-12/sleep.csv 42,984 bytes
health-export-2026-10-12/workouts.csv 10,496 bytes
health-export-2026-10-12/samples.csv 647,810 bytes
health-export-2026-10-12/category-samples.csv 118 bytes
health-export-2026-10-12/nutrition.csv 161 bytesimport pandas as pd
daily = pd.read_parquet("health-export-2026-10-12/daily.parquet")
daily.pivot_table(index="date", columns="provider", values="step_count")How it works
One call per list to GET /v1/users/{id}/exports/{dataset}, which streams the whole list as one
file. The columns, types and time format are described in CSV and Parquet exports.
Why it matters
Your data outlives the apps. An export you hold is a backup if an account closes, a way to move to another app, and the input of any analysis you want to run yourself.
Dashboard
One page with your steps, sleep, resting heart rate and HRV over the last three months, every device together.
Ask an AI about your health data
Give an AI assistant read-only access to your data, then ask it what you want to know: "how did I sleep the weeks I trained the most?", "has my resting heart rate changed since I started running?".