Visual Studio Code v1.109 Changelog
What changed in VS Code v1.109, released 2026-01-01 — explained in plain English.
TLDR
- VS Code is shifting heavily into "multi-agent" development, meaning you can now have multiple AI assistants working on tasks in the background or cloud while you focus on other things.
- You can now "steer" the AI while it's working, queuing up messages or redirecting it if it starts going down the wrong path.
- The AI chat is much smarter: it can ask you clarifying questions, draw flowcharts, and show you exactly what it's "thinking."
- You can now test your apps in a built-in browser right inside the editor.
Want a heads-up when Visual Studio Code updates?
We email you the moment a new Visual Studio Code version drops — plus plain-English release notes like these. No spam.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
VS Code's January 2026 update is all about making the editor a central hub for AI-driven coding. This matters especially to developers who use AI assistants (called "agents") to write or plan code, because it introduces a bunch of new tools to manage, customize, and secure those AI helpers without ever leaving the app.
What's new
- Message steering and queueing: You no longer have to wait for the AI to finish typing to give it new instructions. You can now "queue" a message to send right after it finishes, "steer" it to interrupt and change direction if it's messing up, or stop it entirely to send a new prompt.
- Agent hooks: You can now set up custom background commands (like running a quick script) that automatically trigger at key moments in the AI's lifecycle, such as right before it tries to run a tool.
- Claude compatibility: If you already use Anthropic's Claude AI, you can now reuse your existing Claude configuration files directly in VS Code.
- Ask Questions tool: Instead of guessing what you want, the AI can now pause and ask you multiple-choice or text questions to clarify your instructions before it writes any code.
- Plan agent: A built-in assistant that creates a structured, 4-phase plan before it starts coding, ensuring it actually understands your goals.
- Mermaid diagrams in chat: The AI can now generate interactive flowcharts and diagrams right in the chat window to help explain complex ideas. You can zoom and pan around them.
- Integrated browser: You can now test your web apps in a built-in browser inside VS Code, so you don't have to switch back and forth to another window.
- Kitty keyboard support: Advanced keyboard protocols (often used by power users in the Kitty terminal app) are now available to everyone, making typing in the terminal smoother.
Improvements
- Chat UX: The AI chat feels much faster and more responsive. You can see the AI's "thinking tokens" (the steps it takes to reason through a problem) right alongside the tools it's using, with less visual clutter.
- Agent Session Management: It's much easier to hand off tasks to AI agents running in the background or cloud, check on their progress, and jump in when needed. You can also rename these background sessions to keep them organized.
- Agent Customization: You can now build your own AI workflows, use "Agent Skills" (reusable AI instructions), and share these setups across your whole team or company for consistent results.
- Agent Optimizations: The AI now has "Copilot Memory" to remember context and work smarter. Code searching is also faster thanks to external indexing (a way of searching huge codebases quickly).
- Terminal Enhancements: General quality-of-life updates make the terminal (where you type text commands) smoother and more reliable.
Fixes
- Several security issues were patched to keep your coding environment safe.
- A bug causing issues with the AI chat feature was fixed.
- Various other minor bugs were squashed to improve overall stability.
Breaking changes
No breaking changes in this release.
This is our plain-language summary. Read the complete, official notes on the Visual Studio Code official changelog ↗.
Google Calendar