Issue with HTTP API -- Only works for ~1 day

When I use the HTTP API and deploy the service with the provider stage set to dev, the URL looks like this: https://blablabla.execute-api.us-east-1.amazonaws.com and the stage name for the API is set to $default

After about one day the API will suddenly change so that the url is https://blablabla.execute-api.us-east-1.amazonaws.com/dev without me touching serverless or AWS. The configuration is then screwed up, it has no CORS policy and everything seems to be reset to the defaults except the stage name will then be dev rather than $default.

Any help would be appreciated.

I removed the github continuous integration stuff for this service on the dashboard and it seems to be fine now. I’m guessing there is some weird bug in that system – I’ll just deploy manually from now on.