AIToday
AI Coding AssistantsGitHub Copilot BlogPublished: Sep 11, 2026, 10:00 JST2 min read

GitHub Copilot app adds diff, terminal, browser panels

GitHub Copilot app adds diff, terminal, browser panels

3 Key Points

  1. What happened

    GitHub says the Copilot app now shows a diff panel, a terminal panel, and a browser panel side by side, with additions in green and deletions in red.

  2. Why it matters

    Before, reviewing agent changes meant bouncing between editor, terminal, and browser; now you can accept changes, run npm run dev, test the site, and open a pull request from one place.

  3. What to watch

    The test is whether built-in panels make agent-made changes feel safe enough to merge, since you can prove the code runs before accepting. Watch the Pick & Polish tool for iterating on elements.

WHO IT HITSDevelopers and other non-engineer builders who review AI-generated code in the GitHub Copilot app are the main audience here. It reduces tab-hopping for anyone who has to check what an agent changed, whether it runs, and whether it works before merging.

Ask the AI about this article →

Summaries like this, in your inbox every morning.

Context & Analysis

The post is aimed at beginners using the GitHub Copilot app, and it frames the update as a way to complete the AI coding loop from one window. Previously, checking an agent's work meant moving between an editor, a terminal, and a browser; now those three views are built-in panels in the same app. The article walks through a website example, adding a dev server script that opens the client folder and runs npm run dev, then using the browser panel to test the feature as if using the site.

The diff panel is described as the place to see exactly what an agent changed, with green additions and red deletions, and it lets you accept changes, leave comments, or ask Copilot for edits. The terminal panel is for running the project's own commands, and the browser panel closes the loop for anything with a user interface, including the Pick & Polish tool for selecting an element and adjusting it with the agent. Together they are presented as answering three questions: what changed, does it run, and does it actually work.

The stakes depend on whether seeing review, run, and preview side by side makes agent-made changes feel safe rather than scary for less experienced users. The article presents the panels as a way to prove that what you merge works before you accept it and open a pull request. GitHub Universe 2026 is noted as October 28-29 in San Francisco or online, though the post does not tie the panels to that event.

FAQ
What can I do in the Copilot app without leaving it?
You can review the diff, run commands in the terminal panel, and open the browser panel to test what you built. You can also accept changes and create a pull request directly from the app.
How does the diff panel help me review agent changes?
It shows exactly which lines were added, removed, or changed, with green for additions and red for deletions. From there you can accept changes, leave comments, or ask Copilot to make changes.
Can I run multiple terminal commands at once?
Yes. The article says you can have multiple terminal windows open at once and switch between them. You can run commands by hand or use the Run button for the project's scripts.
GitHub Copilot BlogRead Original Article

Also reported by GitHub Blog (AI)

Get the latest AI Coding Assistants news every morning

For example, today's edition would include:

  • Simon Willison: AI coding agents won't end software engineersSimon Willison's Weblog · 5h ago
  • Cognition has Devin test its own work with GPT-6 AstraOpenAI Blog · 5h ago
  • Nikkei Crosstech NEXT Tokyo 2026 opens, AI-driven development in focusTop Companies AI · 9h ago

AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. Q&As are published on this page for other readers too.

Related Articles

Next articleOracle net income up 60% to $4.679 billion on AI cloud demand