
What happened
The GitHub Copilot app now includes built-in diff, terminal, and browser panels so developers can review, run, and preview agent-generated code in one place.
Why it matters
Previously, checking agent changes required switching between an editor, terminal window, and web browser; now those three steps sit side by side, making agent-made changes easier to verify before merging.
What to watch
The test is whether the built-in loop actually lets teams prove that what they're merging works, as GitHub frames it; watch for whether the panels change how quickly developers accept agent changes.
WHO IT HITSSoftware developers and engineering teams using GitHub Copilot to review and merge agent-generated code can now verify changes without leaving the app.
Summaries like this, every morning.
GitHub is positioning the Copilot app as a single workspace for the full cycle of accepting AI-written code. The company notes that previously, developers had to bounce between an editor, a terminal window, and a web browser to review changes, run the code, and see the result. By embedding those three steps as side-by-side panels, GitHub is betting that the friction of tab-switching is what makes agent-made changes feel risky.
The app now walks a developer through a specific loop: check the diff, start the project from the terminal, open it in the browser, and iterate if needed. GitHub highlights the Pick & Polish tool within the browser panel, which lets a developer select an element and adjust it with the agent — a small feature that keeps the iteration inside the same session. Once satisfied, the developer can accept the change and create a pull request without leaving the app.
What the outcome hinges on is whether this integrated loop actually makes developers more willing to merge agent output. GitHub's stated goal is to replace the anxiety of reviewing AI code with a provable check — did it run, and does it work — and the panels are the mechanism for that proof. Whether teams adopt the app as their primary review surface, rather than continuing to verify elsewhere, is likely to depend on how well the panels handle real projects beyond the website example GitHub describes.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
Cisco released Project Quarry, an open-source engine that evaluates whole AI agent sessions — single- and mult…

In guidelines published October 27, Amazon told developers and managers that consuming more tokens is not the…

IBM introduced a self-hosted deployment option for IBM Bob, its agentic software development platform, deploya…

Corporate legal departments are investing in AI but discovering the biggest obstacle is how their lawyers have…

Schwab Advisor Services made Claude for Financial Advisors available through Schwab Advisor Center, making it…

Palo Alto Networks says its teams used frontier Claude models through Anthropic's Project Glasswing to uncover…
