Allow well-behaved agents to self identify

[mweberxyz]

🏷️ Discussion Type

Product Feedback

💬 Feature/Topic Area

Other

Body

Increasingly, what's posted under my name on GitHub comes from an agent using my token, and it looks exactly like something I wrote. Nobody reading it can tell the difference.

I'd like a standard way for an agent to say "an agent did this," and a marker next to the author's name when it does.

Projects are already inventing their own markers: drivecore/mycoder#368, kornia/kornia#6223, humlab-speech/visible-speech-deployment#259, Adam-S-Daniel/_agent-guidance#319, noahdgrant/nightshift#682. Each uses its own format, and none can mark actions that have no text, like approvals, merges, and closes.

Possible implementations:

• A request header that agents send on every API call
• A setting on the token, chosen when it's created

What this would enable:

• Muting or filtering agents, the way Allow to mute bots asks for bots
• Letting the person mark an agent's post as reviewed
• Pinging someone's agent instead of the person, which cuts down their notifications

Interactive demo: https://mweberxyz.github.io/a-simple-request/

---

I drafted this with an AI agent and reviewed every line before posting. - wouldn't it be nice to not need to say this?

[github-actions[bot]]

💬 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 ⏩

• Your input will be carefully reviewed and cataloged by members of our product teams. 
  • Due to the high volume of submissions, we may not always be able to provide individual responses.
  • Rest assured, your feedback will help chart our course for product improvements.
• Other users may engage with your post, sharing their own perspectives or experiences. 
• GitHub staff may reach out for further clarification or insight. 
  • We may 'Answer' your discussion if there is a current solution, workaround, or roadmap/changelog post related to the feedback.

Where to look to see what's shipping 👀

• Read the Changelog for real-time updates on the latest GitHub features, enhancements, and calls for feedback.
• Explore our Product Roadmap, which details upcoming major releases and initiatives.

What you can do in the meantime 💻

• Upvote and comment on other user feedback Discussions that resonate with you.
• Add more information at any point! Useful details include: use cases, relevant labels, desired outcomes, and any accompanying screenshots.

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. ⭐

[Enzo-Camelo]

+1, this is a real gap and it's only going to get bigger. Every project inventing its own footer ("This was written by an agent") shows the demand, but as you pointed out, text markers can't cover non-text actions (approvals, merges, closes, label changes), and they're trivial to forget or strip.

A few thoughts that might help move this forward:

What's possible today (partial workarounds)

1. Use a GitHub App instead of a PAT. Actions taken with a GitHub App's user access token are already linked to the app: the REST API exposes a performed_via_github_app field on issues, comments and timeline events. That's very close to what you're asking for, it just isn't surfaced clearly in the UI or treated as "agent" specifically.
2. Use a dedicated machine/bot account. This gets you the [bot] badge (if it's an App) and covers approvals and merges, but you lose the "acting on behalf of @user" link, which is the important part of your proposal.
3. Commit trailers. Co-authored-by: or Assisted-by: trailers work for commits, but only for commits.
4. Hidden HTML markers (<!-- agent -->) in comments help automation filter agent posts, but humans can't see them and they don't cover non-text actions either.

None of these give you "this is me, but my agent did it" across all action types, which is why a platform-level feature makes sense.

On the two implementation options

• Token setting (chosen at creation): I think this should be the baseline. It's enforced server side, so it works even if the agent "forgets" to identify itself, and it covers every action the token performs, including approvals and merges.
• Request header: useful as an addition, for example to tell GitHub which agent (name/version) acted, but on its own it's voluntary and easy to omit.

Ideally both: the token marks the action as agent-performed, and an optional header adds detail.

Where the marker should show up

• A small badge next to the author name (e.g. "@user via agent")
• The timeline for non-text events ("@user's agent approved these changes")
• The REST/GraphQL APIs and webhooks, so tools can filter or mute
• The audit log for organizations

That last point matters a lot for orgs that need to know which approvals and merges were actually reviewed by a human.

Great proposal, and the interactive demo makes it very clear. Upvoted.