I received my Creator Micro 2 today and am trying to use it with the Codex desktop app on macOS.
Codex detects the device as connected, but its native controls do not activate. I also cannot add the required ChatGPT layer because the ChatGPT/OpenAI option is missing from the Layers menu in Work Louder Input.
After clicking the “+” button next to Layers in Input, I expect to see an option to add a ChatGPT/Codex layer.
Once that layer is added, the Creator Micro 2 keys should trigger the actions assigned in Codex, such as Fast mode and Push to Talk.
The Layers “+” menu only displays:
Add macOS layer
Add Windows layer
Import layer
There is no ChatGPT/OpenAI option.
The device therefore continues operating as a generic keyboard layer:
The physical key that I assigned as “Fast” in Codex outputs G.
The wide microphone key outputs K and L, because it covers the two physical switches assigned to those letters on the generic layer.
This does not appear to mean that the Codex actions themselves are mapped to G or K/L. Instead, the required Codex/ChatGPT device layer is not active.
Connect the Creator Micro 2 to the Mac.
Open Work Louder Input.
Select the Creator Micro 2.
Click the “+” button next to Layers.
Observe that only the macOS, Windows, and Import options are shown.
Open Codex and assign actions such as Fast and Push to Talk.
Press the corresponding physical keys.
The keys output their generic keyboard letters instead of triggering Codex actions.
Device: Creator Micro 2, Bluetooth model
Device received: Today
Firmware: v0.6.1
Work Louder Input: v0.18.0
Codex desktop app: 26.803.41515
macOS: 26.5.2
Connection methods tested: Bluetooth and USB
Codex/ChatGPT Input Monitoring: Granted
Logi Options+ Input Monitoring: Disabled during testing
Reconnected the Creator Micro 2
Tested both Bluetooth and USB connections
Quit and reopened Work Louder Input
Quit and reopened Codex
Restarted the Mac
Confirmed Input Monitoring permission
Disabled Logi Options+ Input Monitoring
Reconfigured the Creator Micro layout in Codex
The issue remains after all of these steps.
The Work Louder Input log reports:
sys.version WLDeviceError: Request timed out
It then reports:
cannot get fw version of valid device
Could the firmware-version request failure be causing Input to hide the ChatGPT layer option?
Please authenticate to join the conversation.
In Review
Bugs
15 days ago

kkks
Get notified by email when there are changes.
In Review
Bugs
15 days ago

kkks
Get notified by email when there are changes.