Show when an AI agent wrote a PR, and let the human who started it approve it #213014
Replies: 3 comments
|
+1 on the approval part, it's annoying. What works today: run the agent under a GitHub App (or a machine user) instead of your own token. Then the bot is the PR author, you can approve it like any other reviewer, and the audit log shows which identity pushed. One app per team is usually enough, not one per agent. Add a |
|
Welcome to the GitHub Community, @Shreyamitti! You are more likely to get a useful response if you are posting your questions and feedbacks in the applicable category and are explicit about what your project entails--giving a few more details might help someone give you a nudge in the right direction. I've gone ahead and moved it for you. Good luck! |
|
💬 Your Product Feedback Has Been Submitted 🎉 Thank you for taking the time to share your insights with us! Your feedback is invaluable as we build a better GitHub experience for all our users. Here's what you can expect moving forward ⏩
Where to look to see what's shipping 👀
What you can do in the meantime 💻
As a member of the GitHub community, your participation is essential. While we can't promise that every suggestion will be implemented, we want to emphasize that your feedback is instrumental in guiding our decisions and priorities. Thank you once again for your contribution to making GitHub even better! We're grateful for your ongoing support and collaboration in shaping the future of our platform. ⭐ |
Uh oh!
There was an error while loading. Please reload this page.
Introduce yourself
Hi!
Links (optional)
No response
Where are you in your GitHub journey?
Building my first project
And where are you going next on GitHub?
AI coding agents like Claude Code, Cursor and Codex often open PRs using the developer's own credentials. GitHub then lists the developer as the PR author, which causes two problems:
The current workaround is to create a separate GitHub App or bot identity for each agent. That works, but it means managing extra apps, tokens and secrets just to get correct attribution.
What we're asking for:
As more teams have agents write code, being able to tell what humans wrote from what agents wrote, and having a clean review flow for agent work, matters for both governance and speed.
Would others find this useful?
What technical skills or projects are you working on?
AI coding agents like Claude Code, Cursor and Codex often open PRs using the developer's own credentials. GitHub then lists the developer as the PR author, which causes two problems:
The current workaround is to create a separate GitHub App or bot identity for each agent. That works, but it means managing extra apps, tokens and secrets just to get correct attribution.
What we're asking for:
As more teams have agents write code, being able to tell what humans wrote from what agents wrote, and having a clean review flow for agent work, matters for both governance and speed.
Would others find this useful?
Got a question for us? (optional)
No response
All reactions