GitHub Copilot in Visual Studio Code v1.109 – January Release

Source: GitHub Changelog

This January 2026 release brings significant improvements to GitHub Copilot in Visual Studio Code with agent‑driven workflows, enhancements to agent session management, and the introduction of agent support for Claude by Anthropic—right in your IDE.

Agent extensibility (models + integrations)

• Claude agent support is in public preview and enables delegating tasks to Anthropic’s official Claude Agent SDK using Claude models from your GitHub Copilot subscription.
• Richer chat interactions with MCP apps enable more tool‑driven, interactive Copilot experiences inside VS Code.

Copilot Chat & UX

• Faster, snappier chat experience with improved streaming responsiveness.
• Higher‑quality reasoning results for more reliable answers and better task execution.
• Better visibility into model reasoning with thinking tokens and agents that ask questions instead of assuming.
• Revamped editor inline chat to make in‑context Copilot interactions feel more natural and effective while you code.

Multi‑agent development & session management

Improved agent session management makes it easier to delegate work across:

• Copilot
• Claude
• Local agents
• Background agents
• Cloud agents

Seamless handoff and intervention: jump into agent work when needed, without disrupting momentum.

• Agent status indicator shows sessions needing attention; sub‑agents run in parallel for faster completion.

Copilot agent customization

• Agent orchestrations enable building repeatable workflows tailored to your team or project.
• Agent Skills plus org‑wide customization help ensure more consistent outcomes across developers and environments.

Agent optimizations (smarter + faster)

• Copilot Memory helps agents work smarter by retaining relevant context across interactions.
• Faster code search via external indexing improves retrieval speed and responsiveness during large‑repo development.

Agent security & trust

• Terminal command sandboxing increases confidence when agents propose or execute commands (experimental, macOS/Linux only).
• Effective auto‑approval rules improve safety and control for agent‑driven actions while reducing unnecessary prompts.
• Terminal lifecycle improvements give better control over background commands with timeout, await, and kill capabilities.
• Integrated browser: Test and run apps without leaving the editor.
• Terminal enhancements: Quality‑of‑life improvements for smoother, more reliable terminal usage.
• Coding & editor improvements: Multiple refinements to everyday coding flow.
• Extensions & API updates: New capabilities for extension authors to build richer experiences.

Happy coding!

Join the discussion within the GitHub Community.