harden-runner
harden-runner copied to clipboard
Cache endpoint is not auto-detected on first run
As an example, for this workflow, the insights API did not detect the call to the storage endpoint as the cache endpoint. https://app.stepsecurity.io/github/DataDog/stratus-red-team/actions/runs/3156772572
It was detected for the second run.
This is because in the first run, harden runner does not cache a file till the post harden runner step. So, it does not know the cache endpoint in the first run. From the second run, it can access the cache endpoint, due to the file cached in the post harden runner step of previous run.
We should be able to classify the storage endpoint as cache endpoint in the insights API.