❗ [FEATURE REQUEST] ❗- Sub Agent process/billing handling

[KyleC-TIH]

Select Topic Area

Product Feedback

Copilot Feature Area

Copilot Agent Mode

Body

I was told by Clay from GitHub Support to post this here seeing that GitHub works with VS Code for the copilot chat feature.

So after some fairly extensive testing of the agentic handoffs feature that agent mode allows you to define in your .agent.md files, etc. I realized that when an orchestration is being handled, the handoffs to sub agents are handled all within the same context window as your original request, and I'm also not sure if the model definition feature within the .agent.md files is even respected.

The problem with this setup is that users that want to do a sort of long running session (about 2 hours, which is fairly short when you look at Cursor running a week long session to build a shoddy browser) will need to move away from the Copilot Chat feature because it doesn't have the capability to handle these types of orchestration requirements that are becoming more commonplace, especially with the aggressive advertising of agentic capabilities.

So I think Copilot Chat could benefit from having the ability to handle orchestration in such a way that the orchestration agent can spawn a separate thread locally that then handles all of the context and requests for that sub agents needs, with a fully new session being spawned as well (with a nicely discounted rate for sub agents to help give users a reason to choose Copilot over competitors), with only the context deemed necessary by the orchestration agent being passed into that sub agents context.

This would:

• keep sub agent context windows lean and highly focused
• keep orchestration costs encouragingly low enough
• allow true agentic workflows to be achieved since each agent would actually be able to use a specialized model of choice

As it stands, the rate limit is vague, and with no detailed breakdown of how to determine my limit at any given point, the rate limiting is effectively arbitrary to the end user, which makes using copilot chat for multi agent workflows completely unpredictable and highly frustrating. Rate limits should never be based on rules that can't be clearly explained to your end user in such a way that they can monitor their usage accurately if they choose to do so.

I use VS Code every day as a senior dev, and I've been using the Copilot Chat more and more in both a work and personal capacity to the point where I even got a subscription just to get more benefit from the newer models, trying to see how much benefit I can extract from it, and this is a huge pain point in something that is otherwise a great product.

[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. ⭐

[KyleC-TIH]

The 1.109.0 update basically addressed all of these main concerns here. Thanks guys!