Use a pressure mat with a load‑cell sensor, wire it to a small microcontroller, and code a debounce routine that only fires when the weight crosses a set threshold. Add a second sensor in parallel as a check; if either reads high you lock the door. Keep the wiring neat, use shielded cable, and test with a few different weights to calibrate. Put a mechanical override lever in the lock so you can open it if the electronics fail. That’s pretty foolproof and doesn’t need fancy tricks.

Add a tiny vibration sensor to the mat’s mounting plate and use a second load‑cell that only activates if the force changes faster than, say, 200 ms. That catches a heavy object being left there versus a quick step. For the firmware lock, flash the MCU with a one‑time bootloader and disable the programming pins. Keep the lock code in read‑only memory, and use a small physical keypad or NFC tag to reboot into a maintenance mode if you need to update it. It’s a bit of extra wiring, but it keeps the system honest without over‑engineering.

Good call, just tuck the keypad behind a removable panel, use a rolling‑code algorithm, and write the event to a secure EEPROM. Keep the log size small, wrap it in a CRC, and print the hash on a small screen when you open the panel—any tamper shows up immediately. That’s a solid, no‑frills system.

Sure thing—put that tiny screen inside a hidden panel that only pops open when you need to check the log. That way the hash stays invisible unless you’re actually doing maintenance. Stay ahead.