Export Garmin Connect activities to PostgreSQL or CSV

Garmin Connect lets you download your whole account: every activity you ever synced, as the original FIT files. Those files are binary, so a spreadsheet or a database cannot read them as they are. This guide loads a Garmin export into PostgreSQL with Open Health Sync: each activity becomes a workout row with its distance, heart rate and elevation, plus its GPS route and per-second heart rate and power. Then you query it in SQL or export it as CSV or Parquet. No Garmin developer account is needed, and everything runs on your own computer.

1. Request your Garmin exportCopy Anchor Link

1. Sign in at garmin.com/account/datamanagement and open Export Your Data.
2. Choose Request Data Export.
3. Garmin emails a download link, usually within 48 hours, sometimes later.

Keep the zip as it is: the import opens the zips inside it. The JSON files of the export (sleep, stress, daily summaries) are not read; this guide is about activities. To get one activity now, open it in Garmin Connect on the web, then the gear menu, Export Original.

2. Start Postgres and Open Health SyncCopy Anchor Link

You need Docker with Compose v2.24 or later, plus curl, jq and openssl. From a clone of the repository:

git clone https://github.com/soma-intelligence/open-health-sync.git && cd open-health-sync
cp .env.example .env
sed -i.bak "s|^OHS_ENCRYPTION_KEY=.*|OHS_ENCRYPTION_KEY=$(openssl rand -base64 32)|" .env
docker compose --env-file .env -f deploy/docker-compose.yml up -d --wait

Then create your own user, and keep its details in a file the next steps read:

docker compose --env-file .env -f deploy/docker-compose.yml exec -T ohs \
  /ohs admin personal setup > ~/.ohs.env
chmod 600 ~/.ohs.env
set -a; . ~/.ohs.env; set +a

The key in that file reads everything on your instance: treat it like a password.

3. Import the exportCopy Anchor Link

Install the ohs command on your computer: take the archive for your system from the latest release (darwin is macOS), unpack it and put ohs on your PATH. Then import through a token that can only upload:

export OHS_TOKEN=$(curl -s -X POST "$OHS_URL/v1/users/$OHS_USER/tokens" \
  -H "Authorization: Bearer $OHS_KEY" -d '{"scopes":["upload"]}' | jq -r .token)
ohs import files ~/Downloads/garmin-export.zip --dry-run    # counts what it would send
ohs import files ~/Downloads/garmin-export.zip

The command prints counts and formats only, never a file name: Garmin names the files after your email address. Each file is sent once, so running it again, or importing a newer export later, only adds the new activities. --since 2025-01-01 keeps the activities from that day on. The other options are in the activity files guide.

4. Query it with SQLCopy Anchor Link

Open psql in the database container, with your user id in a variable:

docker compose --env-file .env -f deploy/docker-compose.yml exec postgres \
  psql -U ohs -d ohs -v user="$OHS_USER"

Your last activities:

SELECT start_date, activity_type, duration_s / 60 AS minutes,
       round((distance_m / 1000)::numeric, 2) AS km,
       avg_heart_rate, elevation_gain_m
FROM ohs_data.workouts
WHERE user_id = :'user' AND source_slug = 'garmin'
ORDER BY start_date DESC
LIMIT 10;

Running distance, month by month:

SELECT date_trunc('month', start_date)::date AS month,
       count(*) AS runs,
       round((sum(distance_m) / 1000)::numeric, 1) AS km
FROM ohs_data.workouts
WHERE user_id = :'user' AND activity_type = 'running'
GROUP BY month
ORDER BY month DESC
LIMIT 12;

The GPS route of your last ride, point by point:

SELECT r."timestamp", r.latitude, r.longitude, r.altitude_m
FROM ohs_data.route_points r
WHERE r.workout_id = (
  SELECT id FROM ohs_data.workouts
  WHERE user_id = :'user' AND activity_type = 'cycling'
  ORDER BY start_date DESC LIMIT 1)
ORDER BY r."timestamp";

Per-second heart rate and power are in ohs_data.quantity_samples, with the types heart_rate, cycling_power and running_power.

5. Export to CSV or ParquetCopy Anchor Link

curl -sOJ "$OHS_URL/v1/users/$OHS_USER/exports/workouts?format=csv" \
  -H "Authorization: Bearer $OHS_KEY"

format=parquet gives a file for pandas, Polars or DuckDB, and start and end (YYYY-MM-DD) a time window. Each row keeps the file's extra figures (power, cadence, moving time, device name) in its metadata column. Columns and types are described in CSV and Parquet exports; GET /v1/users/{id}/workouts/{workout_id}/route gives one route as JSON.

What the export does not giveCopy Anchor Link

Laps, swim lengths, running dynamics and beat-to-beat intervals stay in the stored FIT file and are not mapped to tables. Sleep, stress and Body Battery are in the export's JSON files, which this import does not read. For those and for new data as it arrives, the Garmin connector uses Garmin's Health API, which needs keys from Garmin's partner program.

NextCopy Anchor Link

• Import other devices' files the same way: COROS, Wahoo, Zwift, Suunto, Polar and Strava archives are covered in the activity files guide.
• Add your other devices: every connector lands in the same tables.
• Ask an AI assistant about your training through the read-only MCP server.