doublemarkpro/esp32-s3-touch-amoled-codexpedometer — explained in plain English
Analysis updated 2026-05-18
Flash a small round touchscreen board to show a smartwatch style clock, weather, and step counter as a desktop gadget.
Monitor OpenAI Codex usage quota and AI coding agent status like idle, working, or error directly on a physical display.
Run the included Python bridge scripts on a Windows PC to feed Codex and Claude Code hook events to the device.
| doublemarkpro/esp32-s3-touch-amoled-codexpedometer | 0xustaz/streamgate | a-bissell/unleash-lite | |
|---|---|---|---|
| Stars | 1 | 1 | 1 |
| Language | Python | Python | Python |
| Setup difficulty | hard | hard | hard |
| Complexity | 4/5 | 4/5 | 4/5 |
| Audience | developer | developer | researcher |
Figures from each repo's GitHub metadata at analysis time.
Requires the ESP-IDF toolchain, flashing firmware over serial, and running separate Python helper scripts on a PC for full functionality.
This repository is built on top of Waveshare's official ESP32-S3-Touch-AMOLED-1.75 development board project, a small circuit board that combines an ESP32-S3 chip with a 1.75 inch round AMOLED touchscreen, touch sensing, motion sensing, a real time clock, audio hardware, and storage support. The repository keeps all of the original examples, schematics, and firmware notes from the board maker, and adds one new example project called CodexPedometer. CodexPedometer turns the round touchscreen board into a small desktop smartwatch style status display. It includes a clock face with an optional simulated hand and time synced over the network, a weather page pulled from a Chinese mapping web service, a pedometer page that uses the board's built in six axis motion sensor to show step count, distance, and estimated calories, and two status pages built specifically for AI coding tools: one shows how much usage quota is left on OpenAI's Codex tool, and another shows whether an AI coding agent such as Codex or Claude Code is currently idle, working, waiting, in an error state, or finished with a task. On first boot the device creates its own Wi-Fi hotspot with a simple web page for entering your home Wi-Fi details, weather city, and the addresses of two small helper programs that run on your computer. Those two helper programs are Python scripts you run on a Windows PC: one reads real Codex usage data from the Codex desktop app and serves it over a local web address, and the other tracks and serves AI agent status, with an option to install itself so it starts automatically when you log into Windows. A Codex hook script and a printable Claude Code hook configuration let those tools report their current state to this status server, which the watch then displays. Getting the CodexPedometer demo running requires the ESP-IDF development toolchain, version 6.0.2 or newer, setting the build target to esp32s3, then building and flashing the firmware to the board over a serial port. The README explicitly warns not to commit personal Wi-Fi passwords, API keys, or private network addresses, since the repository's own default configuration values are placeholders.
A smartwatch style firmware demo for a round AMOLED ESP32-S3 dev board that also shows Codex usage and AI coding agent status.
Mainly Python. The stack also includes ESP32-S3, ESP-IDF, Python.
Setup difficulty is rated hard, with roughly 1h+ to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.