SmartScaleConnect

SmartScaleConnect - application for synchronizing smart scale data between different ecosystems.

Features:

• Load data from Garmin, Zepp Life, Mi Fitness, My TANITA, CSV, JSON
• Save data to Garmin, Home Assistant, CSV, JSON
• Support params: Weight, BMI, Body Fat, Body Water, Bone Mass, Metabolic Age, Muscle Mass, Physique Rating, Visceral Fat, Basal Metabolism, Heart Rate, Skeletal Muscle Mass
• Support multiple users data
• Support scripting language

Inspired by: series of projects from @lswiderski.

Attention. The application is at an early stage of development. The configuration and functionality can change a lot.

Fast start

Download binary.

Configuration

A configuration file with name scaleconnect.yaml may be located in the current working directory or near application binary.

The YAML format is very demanding on indentation and spaces. Please observe them.

After the first launch, the scaleconnect.json file may appear next to the configuration file. It contains the authorization credentials for your services.

Config file example:

sync_alex_fitbit:
  from: fitbit AlexMyFitbitData.zip
  to: garmin alex@gmail.com garmin-password

sync_alex_zepp:
  from: zepp/xiaomi alex@gmail.com xiaomi-password
  to: garmin alex@gmail.com garmin-password
  expr:
    Weight: 'BodyFat == 0 || Date >= date("2024-11-25") ? 0 : Weight'

sync_alex_xiaomi:
  from: xiaomi alex@gmail.com xiaomi-password
  to: garmin alex@gmail.com garmin-password
  expr:
    Weight: 'BodyFat == 0 ? 0 : Weight'
    BodyFat: 'Date >= date("2025-04-01") && Source == "blt.3.1abcdefabcd00" ? 0 : BodyFat'

Garmin

Limitations:

• Currently, two-factor authentication is not supported.
• Currently, China server is not supported.

You can use Garmin data as source:

sync_garmin:
  from: garmin {username} {password}
  to: csv alex_garmin.csv

And as destination:

sync_garmin:
  from: csv alex_garmin.csv
  to: garmin {username} {password}

If you want to upload custom manual data to Garmin, just import it from CSV file.

Xiaomi Mi Fitness

Tested on scales:

• Mi Body Composition Scale S400 (MJTZC01YM) - getting other users data is supported.
• Xiaomi 8-Electrode Body Composition Scale (XMTZC01YM) - getting other users data is not supported yet.

Get data from all scales of the main user:

sync_xiaomi:
  from: xiaomi {username} {password}
  to: csv alex_xiaomi.csv

Get the data of all users from specific scales:

sync_xiaomi:
  from: xiaomi {username} {password} {scales model}
  to: csv all_users_xiaomi.csv

• You can check scales model name from Mi Fitness app > Device > Scale > About device > Device model.
• You can add a filter by username.

Example:

sync_yulia_xiaomi:
  from: xiaomi alex@gmail.com xiaomi-password yunmai.scales.ms103
  to: csv yulia_xiaomi.csv
  expr:
    Weight: 'User == "Yulia" ? Weight : 0'

Zepp Life

Limitations:

• Currently, only login via a Xiaomi account is supported.
• When syncing, you are logged out of the mobile app. It is not known how this problem can be fixed.

Tested on scales:

• Mi Body Composition Scale 2 (XMTZC05HM)

Get data from all scales of the main user:

sync_zepp:
  from: zepp/xiaomi {username} {password}
  to: csv alex_zepp.csv

Getting data from all scales of the selected user:

sync_zepp:
  from: zepp/xiaomi {username} {password} {username}
  to: csv alex_zepp.csv

My TANINA

On Tanita servers, the weighing time is stored with an unknown time zone and may be incorrect.

sync_tanita:
  from: tanita {username} {password}
  to: csv alex_tanita.csv

Fitbit

My Fitbit Aria scales are no longer working. Therefore, I am interested in this service only from the point of view of archived data.

Google bought this service and sent an email with the subject "Take action by February 2, 2026 to keep using Fitbit". There they offer to download all the data, otherwise it will disappear.

Supports retrieving data from the archive:

sync_fitbit:
  from: fitbit MyFitbitData.zip
  to: csv alex_fitbit.csv

CSV

A good format for human-readable text. The time inside the file is always your local time. Well-supported in MS Office. It is convenient to build quick analytics there.

You can use CSV file or HTTP-link as source and CSV file as destination:

sync_csv:
  from: csv http://192.168.1.123/source.csv
  to: csv destination.csv

JSON

Same as CSV, but JSON.

Home Assistant

Each time app starts, the most recent weighing will be uploaded to the Home Assistant.

Modify the example as you like. Add the other parameters of your scales. The data will be saved during HA reboots. Graphs will remember the history of changes.

Add to your configuration.yaml:

input_number:
  alex_weight:
    name: Alex Weight
    min: 50
    max: 80
    unit_of_measurement: kg
    icon: mdi:weight
  alex_bmi:
    name: Alex BMI
    min: 15
    max: 30

Add automation:

alias: Weigth Data
triggers:
  - trigger: webhook
    allowed_methods: [ POST ]
    local_only: true
    webhook_id: "594b7e73-1f0f-4c3c-aded-eeaee78a6790"
conditions: []
actions:
  - action: persistent_notification.create
    data:
      message: "{{ trigger.json }}"  # only for tests
  - action: input_number.set_value
    data:
      value: "{{ trigger.json.Weight|round(2) }}"
    target:
      entity_id: input_number.alex_weight  # change here
  - action: input_number.set_value
    data:
      value: "{{ trigger.json.BMI|round(2) }}"
    target:
      entity_id: input_number.alex_bmi  # change here
mode: single

Add to scaleconnect.yaml:

sync_hass:
  from: xiaomi alex@gmail.com xiaomi-password
  to: json/latest http://192.168.1.123:8123/api/webhook/594b7e73-1f0f-4c3c-aded-eeaee78a6790

Sync logic

Every time you start the app, the weight data is fully synchronized:

• All the source data for the entire time is loaded.
• All destination data for the entire time is loaded.
• If the timestamp completely matches, the data is considered the same.
  • If all other parameters match, the weighting is skipped.
  • If the other parameters are different, the weighting is completely replaced by the new data.
  • If the Weight column is zero, the destination weighting is deleted.
• If the timestamp does not match, the weighing is uploaded to the service.
• If the Weight column has zero, the weighting is skipped.

At the first start, I recommend downloading all your data from all services to CSV file and analyze it in Excel for some incorrect data.

If you need to delete a lot of incorrect data from Garmin, you can download it to CSV file, put zeros in the Weight column, and then upload this CSV file to Garmin again.

Scripting language

You can change the synchronization behavior and change the weighting values using the powerful scripting language - expr.

sync_expr:
  expr:
    Date: 'Date - duration("1h")'  # you can adjust the weighing time
    Weight: 'Weight > 60 ? 0 : Weight'  # you can use conditions
    BMI: 'BMI * 0.95'  # mathematical formulas can be used
    BodyFat: 'BodyFat - 5.0'  # mathematical formulas can be used
    BodyWater: 'BodyWater'
    BoneMass: 'BoneMass'
    MuscleMass: 'MuscleMass'
    SkeletalMuscleMass: 'SkeletalMuscleMass'
    MetabolicAge: 'MetabolicAge'
    PhysiqueRating: 'PhysiqueRating'
    VisceralFat: 'VisceralFat'
    BasalMetabolism: 'BasalMetabolism'
    HeartRate: 'HeartRate'
    User: 'User'
    Source: 'Source + " some other text"'  # adding custom text information

For example, many scales measure the MuscleMass parameter. Although professional scales, including Garmin, measure SkeletalMuscleMass. If you want the MuscleMass parameter to be displayed in Garmin instead of SkeletalMuscleMass, do this:

sync_expr:
  expr:
    SkeletalMuscleMass: 'MuscleMass'  # replace the skeletal mass data with a regular mass

For example, I bought a new scale from Xiaomi and don't want to sync data from the Zepp after a certain date. And also I don't want to synchronize data without fat information (incorrect weighings):

sync_alex_zepp:
  expr:
    Weight: 'BodyFat == 0 || Date >= date("2024-11-25") ? 0 : Weight'

Or I bought a new Xiaomi 8-Electrode Scale in addition to the old one Xiaomi S400. Now the S400 shows a completely wrong fat percentage. But I want to leave the weights from ols scales because they have skeletal muscle mass data, and the new scales don't have this param. Also, I want to ignore weighing without fat information.

sync_alex_xiaomi:
  expr:
    Weight: 'BodyFat == 0 ? 0 : Weight'  # ignore weighing without fat information
    BodyFat: 'Date >= date("2025-04-01") && Source == "blt.3.1abcdefabcd00" ? 0 : BodyFat'  # zero body fat from old scales

Known Scales

Useful links

• https://github.com/lswiderski/WebBodyComposition