{
  "id": 9806768,
  "title": "GitHub Copilot app for Beginners: How to build custom workflows with canvases",
  "url": "https://urgent.news/2026/09/25/github-copilot-app-for-beginners-how-to-build-custom-workflows-with",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-25T18:00:00.000Z",
  "source": {
    "name": "GitHub Blog",
    "slug": "github-blog",
    "url": "https://github.blog/ai-and-ml/github-copilot/github-copilot-app-for-beginners-how-to-build-custom-workflows-with-canvases/"
  },
  "original_language": "en",
  "account": "GitHub Copilot app introductions a novel feature called \"canvases,\" a customizable interface that can be tailored to an individual's specific workflow. Unlike traditional tools that dictate fixed screens, canvases allow users to shape the interface around their desired workflow. A canvas can take many forms, such as a kanban board, release checklist, dashboard, form, or even a spreadsheet.\n\nCreating a canvas is a straightforward process. Simply open an agent session, enter the \"/create-canvas\" skill, and describe your desired workflow in plain English. The prompt should cover three key elements: the workflow the canvas should support, the interface features you want to include, and the agent's capabilities within the canvas. Once the canvas is created, it appears in the right-hand panel of the GitHub Copilot app, ready for use without any coding or design work.\n\nThe beauty of canvases lies in their bidirectional nature. As you and the agent work together, the canvas updates in real-time, allowing for instant collaboration. You can edit buttons, cards, filters, and other controls directly within the interface, and the agent can update the canvas based on its actions. This seamless interaction is akin to using a live shared whiteboard, fostering teamwork and efficient workflow management.\n\nMoreover, canvases are not static. You can refine them over time, adding or removing features as needed. If the initial version doesn't quite meet your needs, you can ask the agent to modify the canvas further, such as adding new columns, filters, or even turning the entire canvas into a checklist for your tasks. The agent will adapt the canvas to your specifications, ensuring it aligns perfectly with your workflow.\n\nOnce a canvas is created, it becomes a reusable extension that can be shared with your team or saved as a personal extension. This enables seamless collaboration, as team members can access and utilize the same customized interface. If you need inspiration, the community has shared various ready-made canvas extensions through the \"Awesome Copilot\" collection, including tools for release notes, kanban boards, issue triage workflows, and more. You can install these extensions and then customize them using the same \"/create-canvas\" skill, tailoring them to your specific needs.\n\nTo get started with canvases, focus on three crucial questions: What information do you want to see, what actions do you want to perform directly, and what should the agent be able to update or perform? If you're unsure where to begin, the GitHub Copilot app is an excellent starting point. Just open a session, run \"/create-canvas,\" and describe a simple board or checklist for a task you're currently working on. This hands-on approach will help you understand how canvases can streamline your workflow and enhance your productivity.",
  "summary": "Describe the interface you need in plain English, then let the agent build a live surface you can both use and update—so you spend less time adapting to tools and more time getting work done. The post GitHub Copilot app for Beginners: How to build custom workflows with canvases appeared first on The GitHub Blog .",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}