GitHub Copilot Automations now learn from outcomes to self-improve prompts

·2026.10.28 09:00

Key point

The feature uses signals like label changes and PR merges to propose prompt refinements, which users must review and approve before application.

1 / 2

Copilot Automations can now analyze real-world outcomes on GitHub to propose improvements to their own prompts, tools, or triggers.
The system creates a self-improvement loop where a secondary automation analyzes user feedback (e.g., label changes, comment acceptance, PR merges) and suggests refinements to the primary automation.
Users retain control over the process; GitHub notifies them of proposed changes, which must be explicitly reviewed and accepted or rejected before being applied.
This capability is available for Copilot Pro, Pro+, Max, Business, and Enterprise plans.

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.