Escape the cloud: Dev hacks Withings scales to dump their server addiction
Tired of your bathroom weight data taking a round trip through the cloud? One dev finally broke the chains of Withings, proving that even a humble scale doesn't need a corporate babysitter to tell you how heavy you are.
Dídac Sabatés decided that his Withings Body+ scale didn't actually need to chat with the company's servers to function. By using an ESP32 microcontroller and a custom component built for ESPHome, he managed to intercept the data flow directly. Normally, the device relies on a proprietary API that forces every single measurement through the manufacturer's cloud, which acts like a digital gatekeeper for your own body fat percentage.
To bypass this, Sabatés reverse-engineered the closed Bluetooth protocol. He fed the decompiled code of the older Android app into Codex to interpret the handshake logic. The process involved training the AI to recognize how the scale broadcasts measurements after identifying the specific user ID. Instead of waiting for a corporate server to cough up the data, the ESP32 now grabs the packets via Bluetooth and shoots them straight into Home Assistant over the local network.
The integration captures more than just the basic weight; it pulls in bone mass, water percentage, and muscle data effortlessly. By using template sensors in Home Assistant, the information persists even when the microcontroller is offline, ensuring the dashboard doesn't go blank during a reboot. This is less about high-level hacking and more about reclaiming the right to own the metrics generated by your own feet.
It is genuinely amusing to see a piece of hardware that costs a premium essentially functioning as a glorified data-leaking machine. The fact that an AI had to be called in just to stop a scale from calling home proves that the "smart" home is often just a fancy term for a remote-controlled surveillance kiosk.
Source: Dídac Sabatés
Comments
This is where the magic happens: AI reads your discussion and rewrites the article based on the most interesting comments. Each strong comment adds points to the meter below. Once the meter is full, the article updates live — no page reload needed.