How to build custom workflows with canvases in the GitHub Copilot app | Tutorial for beginners

Name: How to build custom workflows with canvases in the GitHub Copilot app | Tutorial for beginners
Uploaded: 2026-09-01T15:01:24+00:00
Description: GitHub introduces canvases in the GitHub Copilot app and shows how beginners can use the /create-canvas skill to generate custom, interactive workflow...

Today by GitHub

GitHub introduces canvases in the GitHub Copilot app and shows how beginners can use the /create-canvas skill to generate custom, interactive workflow surfaces like release checklists, kanban boards, and triage views, then iterate on them and share state in real time.

Overview

This tutorial explains how canvases in the GitHub Copilot app provide customizable, bidirectional surfaces that you and AI agents can work on together.

What a canvas is in the GitHub Copilot app

A canvas is a workspace interface you can describe in plain English.
The Copilot app can generate the canvas for you, and you can continue collaborating with it after creation.

Bidirectional collaboration

The canvas is designed to be bidirectional, meaning both you and the agent can update and refine the same surface.
Changes can be shared with others through real-time state sharing.

Building a canvas with `/create-canvas`

Use the /create-canvas skill to generate a canvas from a natural-language description.
Example canvas types mentioned:
- Release checklists
- Kanban boards
- Triage surfaces

Iterating and refining

After the initial canvas is created, you can iterate on it by refining the prompt and adjusting the interface.
The focus is on quickly shaping the canvas into a workflow that matches your needs.

Using community canvases

The video points to community canvases available via Awesome Copilot.
These can be used as starting points for your own workflows.

Canvases can be saved.
The app supports real-time state sharing so others can collaborate on the same canvas state.