# How to Continue a GitHub Copilot App Session in VS Code

Seamlessly transition AI-generated code sessions from the Copilot app into your local development environment.

- **Source:** [The GitHub Blog](https://brifing.si/source/9783c41a-41dd-45e0-8c71-c0930937342d)
- **Published:** 2026-10-19
- **Topic:** [Software & Engineering](https://brifing.si/c/engineering)
- **Original article:** <https://github.blog/ai-and-ml/github-copilot/how-to-continue-a-github-copilot-app-session-in-vs-code/>
- **On Brifing:** https://brifing.si/s/507254e0-7192-4f47-a209-1bc34db7c3a8

## Bridging AI and Editor

Developers often start complex tasks in the GitHub Copilot app but require the robust debugging tools of Visual Studio Code to finalize work. You can now open active sessions directly in your editor. This integration preserves the original task context, conversation history, and the specific worktree, eliminating manual file transfers.

## Maintaining Session Context

When you select Open in VS Code from the Copilot app, the IDE loads the exact branch and worktree associated with the session. The Copilot sidebar retains the full chat history and plan. This allows you to query the agent about specific implementation decisions without needing to re-explain the project requirements.

## Reviewing Agent Changes

Before finalizing code, use the Source Control view to inspect diffs line by line. Verify that the agent followed repository conventions and handled edge cases correctly. If any implementation detail remains unclear, you can continue prompting the agent within the sidebar to explain its logic or request specific modifications to the code.

## Manual Code Refinement

After the agent completes the bulk of the work, you can perform hands-on edits directly in VS Code. This is ideal for minor adjustments like renaming variables, refining error messages, or applying specific styling conventions. These manual changes build directly upon the agent's work within the existing, isolated working tree.

## Local Validation Workflow

Utilize the VS Code integrated terminal to run standard validation commands such as npm test or npm run lint. Treat the agent-generated code with the same rigor as human-written code. Verify UI changes across different viewport sizes and ensure that all loading, empty, and error states function as expected locally.

## Iterative Development Cycles

You can alternate between automated agent assistance and manual editing throughout the session. For instance, after making a manual UI adjustment, ask Copilot to generate a corresponding test case. Because the session context remains active, the agent understands your recent manual changes and can provide relevant, context-aware code suggestions.

## Streamlined Pull Request Management

Once testing is complete, commit and push your changes using standard Git commands within the terminal or Source Control view. Since you are working on the branch associated with the original session, your updates automatically sync with the existing pull request. This avoids the need for manual file copying or reconnection.

## Optimizing Developer Productivity

This workflow creates a clean division of labor between AI and human developers. Copilot handles repetitive implementation tasks, while you retain control for detailed review and final polish. By keeping the session, diff, and worktree connected, you ensure a continuous, efficient development cycle from initial prompt to final code deployment.

## Sources covering this story

- https://github.blog/ai-and-ml/github-copilot/how-to-continue-a-github-copilot-app-session-in-vs-code/ — 

---

_Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._