I am somewhat suspicious that it thinks that was my first PR in my fork. So opened a couple more and it commented on those too. I note that in the action's own repo the dependabot PRs are getting flagged for the new contributor message actions/first-interaction#363 (comment)

Something seems off to me.

The problem that this PR was trying to fix is now tracked at actions/first-interaction#365. The problem of every PR being identified as a new contribution is now tracked at actions/first-interaction#369. So I think it's better to wait and see what happens over there.

Replacement for #30633, since if you force push when the PR is closed, you can't re-open.

Off-topic: You can reopen a closed pull request by first force-pushing the expected commit hash, reopening and then force-pushing the wanted commit hash again: https://gist.github.com/robertpainsi/2c42c15f1ce6dab03a0675348edd4e2c.

I have disabled the workflow (via Github Web UI) in the mean time until the problems are addressed (As it is not doing what we intend and is giving CI failures where they are not warranted)

@rcomer would you like me to take a look? I might potentially come up with a different solution so wanted to check with you first. Thanks!

@melissawm feel free to look. I’m not actively doing anything on this at the moment. I am wondering if we might need to use some other bot. xarray uses https://github.com/apps/welcome and I’m sure there are others.

@bjlittle recommends https://github.com/plbstl/first-contribution

I think that worked @rcomer: melissawm#30

I can't push the changes here, but you can see them in my PR there: https://github.com/melissawm/matplotlib/pull/30/changes

The issue label also works 🎉 melissawm#32

@melissawm do you want to open a new PR and we close this one?

Sure, here it is: #31090