GitHub publishes Copilot agent engineering workflow guide
What happened
On November 7, GitHub published a guide for agent-driven development with Copilot, extending AI-assisted work from code generation to four stages: planning, building, verifying and observing after release. The guide says teams should first gather context and write issues with acceptance criteria, then work the tasks in isolated sessions. Next they split dependent pull requests, run quality and security checks, and turn post-release runtime evidence into follow-up tasks. It cites Faros AI research finding that pull requests spent 441% more time in review after teams moved from low to high AI adoption.
Written by AI from the coverage · updated 4 hours ago
Coverage
Follow the reports to see the story from different sides.
- GitHub Blog · AI & MLBeyond code generation: A developer’s guide to agentic engineering
GitHub 给出 Copilot 智能体工程指南，将 AI 辅助开发扩展到规划、构建、验证与上线观察四个阶段。流程涵盖汇集上下文并生成带验收标准的 Issues、在隔离会话中推进任务、拆分依赖关联的拉取请求并执行质量与安全检查，以及利用运行证据形成后续任务。文中引用 Faros AI 研究称，团队从低 AI 采用率转向高采用率后，拉取请求在评审环节停留的时间增加了 441%。
Heat over time
Heat now 9·Comparable peak 10（Nov 7 07:00）·Comparable change over 24 hours –
The trend only compares accounts observed without gaps, so its range may be smaller than the current heat. Hover or tap the chart for each hour; the left and right arrow keys step through it.