Copilot CLI が大幅刷新——ラバーダック・プロンプトスケジューリング・音声入力が正式提供 Copilot CLI: Improved UI, rubber duck, prompt scheduling, and voice input

GitHub Copilot CLI is receiving one of its most significant overhauls to date, timed to coincide with Microsoft Build 2026. The update signals a clear intent to embed AI more deeply into everyday terminal workflows rather than treating the CLI as a lightweight add-on to the broader Copilot ecosystem.

Three features have reached general availability with this release. The rubber duck feature draws its name from the well-known debugging technique popularized in Andy Hunt and Dave Thomas's 1999 classic The Pragmatic Programmer, where a developer explains a problem aloud—traditionally to a rubber duck—to surface their own blind spots. Copilot now takes on that conversational role, providing an interactive sounding board for working through tricky bugs or design decisions. Prompt scheduling allows developers to define prompts that execute automatically at specified times or under certain conditions, opening the door to lightweight automation of repetitive tasks directly from the terminal. Voice input rounds out the GA trio, letting users speak naturally to Copilot rather than type, which has meaningful implications for accessibility and for scenarios where hands-free interaction matters.

An experimental new terminal interface is also part of the refresh. Details from the changelog snippet are limited, but the framing suggests a visual and structural rethinking of how Copilot surfaces information and controls within the terminal environment—potentially moving beyond a pure text-exchange model.

The timing of this release is telling. The AI coding assistant space has grown intensely competitive over the past year, with tools built on Anthropic's Claude, purpose-built editors like Cursor, and integrated offerings from JetBrains all vying for developer attention. GitHub's strategic response is to deepen its foothold in the terminal, an environment that remains central to how professional developers work regardless of which editor or cloud platform they prefer.

Prompt scheduling in particular could prove attractive to DevOps-oriented teams. The ability to schedule AI-driven prompts alongside existing automation scripts suggests potential integration points with CI/CD pipelines and infrastructure workflows, though the full scope of what's possible will depend on implementation details not yet fully disclosed.

Voice input is worth noting from an accessibility standpoint as well. For developers with motor disabilities or repetitive strain injuries, reducing keyboard dependency in a CLI context is genuinely meaningful—a dimension that often gets overlooked in feature announcements focused on productivity metrics.

Overall, this update positions Copilot CLI as a more autonomous and flexible tool rather than simply a command-completion layer. Whether the experimental terminal interface accelerates or complicates adoption will likely depend on how well it integrates with existing terminal emulators and shell configurations that developers already rely on.