Skip to main content

Background agents in GitHub Copilot

A while ago I blogged about how we could setup our own multi-agent workflow by using Git worktrees. With Git worktrees you can setup a branch as a separate folder. If you combine this with the GitHub Copilot CLI you could have multiple agents all working simultaneously on multiple branches while you can continue development as well.

Although the process I described works quite well, it remained somewhat cumbersome and was not well integrated into the VSCode development experience. Not anymore with background agents…

The problem with foreground-only agents

By defaullt, the Copilot agent operates in the foreground. When you ask Copilot to refactor a function or write tests, you're essentially blocked until the task completes. Your editor is locked into that conversation, that context, that task.

This works fine for quick queries, but what about:

  • Running comprehensive test suites while you continue development
  • Refactoring an entire module while you work on a new feature
  • Analyzing code quality across your codebase while you fix a bug
  • Generating documentation while you implement new functionality

These are tasks that could happen in parallel, but most coding assistants force them to be sequential.

How background agents change this

Background agents flip this model. Instead of blocking your primary workflow, you can delegate tasks that run asynchronously. This isn't just about multitasking—it's about matching the cognitive model of how we actually work. We constantly context-switch between different tasks at different levels of abstraction. Background agents make AI assistance fit that natural workflow.

A background agent can work in its own worktree, running tests, making changes, and committing code—all while you work in your primary workspace. When it's done, you can review its changes and merge them in.

Creating a background agent

There are multiple ways to create a background agent in VS Code. Let me show you one way…

In VSCode go to the Chat window and click on the + sign. Choose New Background agent from the dropdown menu:

You can choose to run the background agent directly in your workspace or give it a separate worktree:

That’s all! You no longer need to create a worktree yourself, but VS Code will handle all the work for you.

You can follow up the active agent sessions (foreground and background) using the sessions view:

 

Remark: As shown in my post yesterday you can also switch to a background agent during a handoff.

More information

Git worktrees–A first step towards a multi-agent development workflow

Github Copilot on the command line

Background agents in Visual Studio Code

Popular posts from this blog

Podman– Command execution failed with exit code 125

After updating WSL on one of the developer machines, Podman failed to work. When we took a look through Podman Desktop, we noticed that Podman had stopped running and returned the following error message: Error: Command execution failed with exit code 125 Here are the steps we tried to fix the issue: We started by running podman info to get some extra details on what could be wrong: >podman info OS: windows/amd64 provider: wsl version: 5.3.1 Cannot connect to Podman. Please verify your connection to the Linux system using `podman system connection list`, or try `podman machine init` and `podman machine start` to manage a new Linux VM Error: unable to connect to Podman socket: failed to connect: dial tcp 127.0.0.1:2655: connectex: No connection could be made because the target machine actively refused it. That makes sense as the podman VM was not running. Let’s check the VM: >podman machine list NAME         ...

Cache stampede: when our cache turned against us

While investigating some performance issues, we ran into an ASP.NET Core API that cached a fairly expensive aggregation query for 60 seconds. Under normal load, that was fine: one request rebuilds the cache, everyone else reads from it. Under peak load, dozens of requests would arrive in that same expiry window, all see a cache miss, and all fire the same expensive query in parallel. The database didn't like that. That was the moment when our caching layer stopped helping and started hurting. A burst of requests comes in at the same time, all miss the cache, and all go hammer the database or the downstream API at once. That's a cache stampede . The cache was supposed to protect our backend, and for a few hundred milliseconds it did the opposite. Why this happens IMemoryCache.GetOrCreate (and its async sibling) looks like it protects you, but it doesn't add any locking on its own. Look at the naive version: public async Task<Report> GetReportAsync(string key) ...

Azure DevOps/ GitHub emoji

I’m really bad at remembering emoji’s. So here is cheat sheet with all emoji’s that can be used in tools that support the github emoji markdown markup: All credits go to rcaviers who created this list.