I think that's perfectly fine to upstream, provided there's understanding and you're communicating as yourself. Having Claude generate all the communication for you without your input makes it really, really hard to tell who you are. For some smaller packages, the review may be small and "LGTM". For others, like Django, there's going to be a few rounds of back and forth, and for that a reviewer shouldn't have to communicate with a bot without their consent. Does that make things a bit clearer?
I appreciate the compliment. I'm not sure. I've got a few PRs that I've been putting off reviewing because they're either an Agent or someone copy-pasting the output without review. There's an investment I'm not ready to pay quite yet, but I suspect I'll include a link to this when I respond.
Hey, Tim here. OP and one of the admins for Djangonaut Space. I know it's demoralizing to get multiple rejections. We do value people who apply multiple times and have a better system for highlighting that now. I hope you reconsider again in the future. Or if you can, go to a DjangoCon and stay for the sprints. Those tend to provide a similar energy.
