Copilot: General Cloud Usage
Learn to use GitHub Copilot from the cloud (github.com)
By the end, participants will:
Know how to chat with Copilot and provide relevant context
Understand how to bundle code, assets, and chats in Spaces
Be able to use Copilot’s Agent mode to make changes to the codebase
Learn how to assign GitHub issues to Copilot
Understand how to configure Copilot’s settings
Duration: ~45 mins
Audience: Engineers, technical professionals and TPOs
👉 Bitovi can help you integrate this into your own SDLC workflow: AI for Software Teams
Introduction
Cloud Copilot is GitHub Copilot built right into GitHub.com. You can use it in your browser through Copilot Chat, Spaces, and Agents. It works alongside your repos, issues, and pull requests, using GitHub-native context like files, folders, multiple repos, and saved conversations.
Why Use Copilot in the Cloud?
Zero setup: Works anywhere with a browser — no local toolchain needed.
Full context: Copilot can see multiple repos, issues, and files at once for more complete answers.
Shared brains: Save and reuse conversations, instructions, and contexts across your team.
From chat to code: Agents can create branches, open PRs, and iterate — all in the cloud.
Perfect for reviews & onboarding: Summarize repos, trace PRs, or draft issues without cloning.
Governance built in: Policies, exclusions, and instruction files live right in the repo.
How It Differs from Copilot in VS Code
Cloud Copilot and Copilot in VS Code share the same AI core but serve very different purposes.
Cloud Copilot is all about understanding, planning, and collaborating at the repository level. It can see across multiple repos, issues, pull requests, and even past conversations — giving it a broader view of your project. You interact with it through chat, Spaces, and Agents right in your browser, no setup required. When it makes changes, it does so by creating branches and pull requests in the cloud, so everything stays auditable and shareable.
Copilot in VS Code, on the other hand, is your in-editor assistant. It’s strongest when you’re writing or refactoring code in files you already have open. It works locally, completing code inline, suggesting edits, and speeding up day-to-day coding tasks. You handle commits and pushes yourself from your IDE.
A few key differences:
Setup: Cloud Copilot runs in the browser — no extensions or SDKs needed. VS Code requires the Copilot extension and a local toolchain.
Context: Cloud Copilot reasons over multiple repos, issues, and PRs; VS Code Copilot focuses on your open workspace.
Collaboration: Cloud Copilot saves and shares conversations, instructions, and Spaces across your team; VS Code Copilot’s context is personal and local.
Change flow: In the cloud, Agents can open PRs and iterate autonomously. In the IDE, you’re driving the edits directly.
Think of it this way:
Use Cloud Copilot to understand, plan, and coordinate across projects.
Use VS Code Copilot to write and refine code quickly in your local environment.
Training Overview
This workshop will walk you through the core features of Cloud Copilot, how it integrates with GitHub, helps you explore and understand repositories, and enables you to collaborate and automate directly in the browser.
You’ll get hands-on experience using Chat, Spaces, and Agents, and learn how to assign issues, adjust repo settings, and make changes — all without ever leaving GitHub.
How the Session Works
You’ll start by choosing a repository you’d like to work with — ideally something familiar or relevant to your team. Throughout the session, you’ll use Cloud Copilot to:
Explore and summarize your repo’s structure and dependencies
Ask questions and reason about issues or pull requests
Draft and refine issues, docs, or small code changes
Use Agents to propose and iterate on pull requests
Experiment with Spaces to organize context and share workflows
Review settings that control Copilot’s scope, rules, and governance
Each exercise builds on the last, showing how Cloud Copilot fits into your daily GitHub workflow — from quick insights to automated changes.
✏️ Setup
To follow along with this training, you’ll need a few things in place:
1. Repo & use case
Pick a repository on GitHub that you’ll use for the session — ideally one you’re familiar with or want to explore further.
Think ahead to a feature or addition you’d like the AI-assistant to help with during the workshop.
2. Copilot access
Ensure you have access to GitHub Copilot. This can be a personal subscription (e.g., Copilot Pro) or via your organization (Copilot Business/Enterprise). GitHub Docs+2GitHub Docs+2
If you’re using an org account: check that your organization has enabled Copilot seats and that you are assigned one. GitHub Docs
If you’re using a free tier, you can still participate, though some advanced features might be limited. GitHub Docs
3. Browser access
Since this training uses the cloud version of Copilot (on http://GitHub.com ), make sure you can access GitHub from your browser — no local IDE or toolchain setup required.
Cloud Copilot Features
In this section, we’ll outline the core cloud Copilot features as you inspect, make changes to, and bundle context from your chosen repository. Read through each document and complete the exercises at the end. By the time you’re done you’ll have a full understanding of how to use Copilot on the cloud!
Have conversations with Copilot entirely on the web.
Bundle context across repositories, issues, PRs, and code.
Use Copilot to make changes to your repo
Assign issues to Copilot and watch as it automatically works on them
Learn how to configure Copilot and set up MCP servers
Conclusion
You’ve seen how Cloud Copilot brings GitHub Copilot into your browser, helping you explore, plan, and automate directly inside your repos.
Today, you learned how to chat with Copilot to understand and navigate your codebase, create and share Spaces for richer context, use Agents to draft issues and open pull requests, and adjust repo settings to guide Copilot’s behavior and governance.
🚀 Going Forward
Use Cloud Copilot for daily repo exploration, reviews, and planning
Share Spaces and instructions with your team to build shared knowledge
Let Agents handle small fixes or documentation updates
Keep experimenting — Copilot gets better as your prompts and context do
Bottom line: Cloud Copilot turns GitHub into a space where AI helps you understand, plan, and ship — all in the cloud.