From CLI to PR: Automating the path to merged code | BRK203

Cassidy Williams and Evan Boyle demonstrate an end-to-end, agentic GitHub Copilot workflow that starts in the terminal and ends with a merged pull request, focusing on the practical mechanics teams need to make AI-assisted sprint work repeatable.

Overview

The session walks through applying GitHub Copilot as an “agentic partner” across a day-to-day engineering loop: planning from the CLI, delegating work, and automating parts of pull request review.

Key areas covered include:

• Using GitHub Copilot in the terminal as part of sprint execution (not just chat-based assistance)
• Context management patterns for keeping AI tasks grounded in the right repo/work scope
• Advanced GitHub Copilot CLI capabilities, including:
  • New commands for research and experimental features
  • Voice mode and speech-to-text input in the CLI
  • An /every automation command for scheduling tasks
• Integrating Spec Kit with “Plan Mode” to structure work before execution
• Controlling cost/throughput with token budgeting and efficiency controls
• Discussion of failure modes and behavior issues (including an audience-raised “goblin mode” personality/behavior concern)

Session resources

• https://aka.ms/build26/BRK203