Post

“the attacker used an orphaned commit to gain access to the workflow run that stores the OIDC token, effectively bypassing the project’s existing publishing protections” well, wait: an external fork had access to the OIDC token? how?

Probably pr target trigger again 😬 It runs the action in the context of the source repo but using the code from the fork. Tough one to keep secure

More or less; yes. We'll have more information in a blog post, but for now we're locking things down and fixing the problem. We have a good understanding of how this attack vector was carried about

This attack is almost exactly the one I had ready to demo the on stage for JS Conf last year before working with @reggi.com. They fixed the ability to publish from this trigger but looks like that missed fix the token exfiltration (that @darcyclarke.me demo’d at the collab summit the same week).

I am serious when I say we have been fighting this battle since mid last year. And people just weren’t listening. Example: pypi got this report and decided it was not urgent to fix. I don’t think they liked how I was pushing on the issue either, but they had the same issues in TP for pypi.

But we couldn’t exactly talk about publicly since it was ambiguous if it was a vuln and we would be the baddies for mishandling a report. So we told them all about it and then waited.

Obviously that has changed now that this has happened and I guess we can be a bit more forward about why we were so opposed to widespread adoption of trusted publishing.