Reviewing agent-generated code usually means hopping between tabs. Here’s how to view diffs, run terminal commands, and preview web apps side by side in the GitHub Copilot app.
Kayla Cinnamon@cinnamon-msft
3 minutes
-
Share:
When an agent changes your code, you need to review it, run it, and confirm it did what you asked. The GitHub Copilot app lets you do all three without leaving the window.
Previously, that meant bouncing between your editor, a terminal window, and a web browser. With the diff, terminal, and browser panels built into the Copilot app, checking an agent’s work happens in one place.
Below is a walkthrough of each panel and how they fit together to complete the AI coding loop.
Using the diff panel to review changes
A diff is a before-and-after comparison, and the diff panel shows exactly which lines were added, removed, or changed—additions highlighted in green, deletions in red. That clarity lets you decide what happens next: accept the changes, leave comments, or ask Copilot to revise them. You stay in control and make the final call.
Running commands in the terminal panel
Reading code is good; running it is better. The terminal panel lets you run commands inside the session without leaving the Copilot app. If that sounds intimidating, it usually isn’t—you’re mostly running the project’s own commands and reading the output.
You can run code by hand or configure it as a script, available through the Run button. Take a website project as an example:
- Add the
dev serverscript that opens the client folder and then runsnpm run dev. - Click Run to start the server for the website.
Multiple terminal windows can stay open at once, so you can switch between them and keep running commands.
Using the pick & polish tool in the browser panel
For anything with a user interface, the browser panel closes the loop. In the website example, that means opening the panel and testing your new feature as if you were using the live site.
To keep iterating, use the Pick & Polish tool to select an element and adjust it with the agent.
To verify what got fixed, run the dev server script again after making changes.
Closing the loop of reviewing AI code
At this point you’ve reviewed the diff, started the project in the terminal, checked it in the browser, and iterated until the feature works. Once it’s where you want it, accept the change directly from the Copilot app and create a pull request. Finishing the loop in one place means no tab hopping, no switching apps, and no losing your place.
Review, run, and preview side by side is what makes agent-made changes feel safe rather than risky—you can prove that what you’re merging actually works.
Take this with you
These built-in panels answer three questions worth asking before you accept any agent’s work.
- What’s changed?
- Does it run?
- And does it actually work?
Working through that list before accepting agent-generated code helps you understand what changed and whether it behaves the way you want—keeping you in control. ✨
Try using the GitHub Copilot app >
Tags:
Written by
Related posts
Project HydraFusion: Frontier quality via multi-model orchestration
In controlled offline evaluations, HydraFusion’s selective coding workflows matched or exceeded the evaluated Opus 5 baseline while reducing estimated workflow cost. Now available as a research preview in GitHub Copilot.
GitHub Copilot app for Beginners: Run several agents at once
Learn how to run parallel agents in the GitHub Copilot app, and experience the moment it stops feeling scary and starts feeling powerful.
Decoding the new AI lingo: Loops, harnesses, squads, hill climbing… oh my!
From loop engineering to harnesses, squads, and open weights, the GitHub Podcast breaks down the AI terms showing up in developer conversations.

