Stop Restarting the Chat: Continue a GitHub Copilot App Session in VS Code
Last Updated on November 2, 2026 by Editorial Team
Author(s): Dave R | Microsoft Azure & AI MVP ☁️
Originally published on Towards AI.
The conversation, diff, and git worktree all carry over. How it works under the hood, where it can catch you out, and a seven step review flow before you push.
Your Copilot agent finished most of the change. Now you need to read the diff, fix a label, and run the tests, and that work happens in VS Code. You don’t need a new chat to get there. Select Open in VS Code on a GitHub Copilot app session and VS Code opens the same git worktree with the conversation, plan, and diff intact. Pushing from VS Code updates the pull request already tracking the work. This article starts with the workflow, then explains the architecture behind it and the five edge cases to know before you rely on it.
The rest of the article walks through a seven-step process for continuing a GitHub Copilot app session in VS Code—inspect and potentially fix the plan in the app, open the session in VS Code, review and test like a pull request, ask the agent targeted questions, edit quickly by hand when faster, run the real checks in a clean worktree, then commit and push through the existing PR. It also covers practical guidance on when to prompt vs. type, what the underlying “handoff” architecture looks like (shared session state across clients, an Agent Host as the source of truth, and shared worktrees/branches), and several gotchas that can trip teams up (ignored files not following, worktrees isolating files but not services, not moving worktree folders, forks not being branches for isolation, and delete behavior). Finally, it notes team governance expectations and concludes with a workflow habit: let the agent draft, then review and verify in VS Code using the same session context.
Read the full blog for free on Medium.
Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.
Published via Towards AI
Towards AI Academy
We Build Enterprise-Grade AI. We'll Teach You to Master It Too.
15 engineers. 100,000+ students. Towards AI Academy teaches what actually survives production.
Start free — no commitment:
→ AI Engineering For Production — our new book, coming before the end of 2026. Join the waitlist.
→ Agents Architecture Cheatsheet — 3 years of architecture decisions in 6 pages
Our courses:
→ AI Engineering Certification — 90+ lessons from project selection to deployed product. The most comprehensive practical LLM course out there.
→ Agent Engineering Course — Hands on with production agent architectures, memory, routing, and eval frameworks — built from real enterprise engagements.
→ AI for Work — Understand, evaluate, and apply AI for complex work tasks.
Note: Article content contains the views of the contributing authors and not Towards AI.
Related posts
Recent Posts
![[Day 37/100] Build Your Own Agent Memory Layer with Postgres and pgvector](https://miro.medium.com/v2/resize:fit:700/1*-Doos2nmUT3FPZ-Qxbk3Hw.png "[Day 37/100] Build Your Own Agent Memory Layer with Postgres and pgvector")
