VS Code 1.141

[Deprecated features and settings] GitHub Enterprise URI setting: The github-enterprise.uri setting is deprecated in favor of github-enterprise.uris , which accepts multiple GHE.com and GitHub Enterprise Server instances. VS Code still reads github-enterprise.uri when github-enterprise.uris isn't configured.

[Agent security] Sandboxing in the Copilot agent host: Settings: chat.agent.sandbox.enabled Sandboxing gives you more control over what Copilot can access while working on a task. It limits how supported agent operations access files and network resources, and helps to reduce the impact of model mistakes, prompt injection, untrusted dependencies, and locally launched tool servers. Depending on the operation and platform, these restrictions use operating-system protections or checks within the agent host process. Sandboxing adds a layer of protection, but does not replace endpoint security or provide a standalone security boundary. Sandboxing is available on Windows, macOS, and Linux. On Windows, required operating-system updates must be installed. Refer to Check platform availability for more details. To configure sandboxing for all your sessions, enable the chat.agent.sandbox.enabled setting. You can

[Agent loop] Copilot harness: The Copilot harness adds exciting new agent functionality to VS Code. It's powered by the Copilot SDK, so its behavior and capabilities are consistent with other Copilot products, including the standalone GitHub Copilot app and the Copilot CLI. To start using the Copilot harness, select it from the harness picker in the chat input. In this release, it might already be the default selection for you. You can continue working as you normally do. The harness runs in a dedicated agent host process based on the Agent Host Protocol (AHP), so you can connect to the same agent session from multiple VS Code windows. Learn how to work with the Copilot harness, or explore the architecture and workflows in the agent host blog post. If you have feedback or requests, please file an issue.

[Authentication] Sign in to multiple GitHub Enterprise instances: Setting: github-enterprise.uris Many organizations use GitHub Copilot through a GHE.com account but keep their code on GitHub Enterprise Server (GHES). The github-enterprise.uri setting held a single instance, so you couldn't be signed in to both at once. For example, you couldn't use Copilot with your GHE.com account and review pull requests on GHES in the same VS Code window. The new github-enterprise.uris setting takes a list of GHE.com and GHES instances: The GitHub Enterprise authentication provider holds accounts from every instance in the list. Each account is labeled with its host, such as monalisa - octocat.ghe.com, so you can tell them apart in the Accounts menu. When more than one instance is configured, signing in asks which instance to use. The order of the entries doesn't select a default instance. To check or change which

[Agent loop] Reclaim storage from inactive worktrees: Run Chat: Open Worktree Cleanup to see how much disk space inactive agent session worktrees use and remove the ones you no longer need. Filter sessions by how long they have been inactive, review each worktree's size, and select the worktrees to clean up. The cleanup editor also lets you configure automatic cleanup for sessions with merged pull requests and control whether VS Code suggests cleanup when session storage grows large. When chat.agentSessions.sessionStorageCleanupSuggestion.enabled is enabled, the Agents window notifies you when enough inactive worktrees accumulate or they use enough disk space to make cleanup worthwhile. Cleaning up marks the selected sessions as done and deletes their worktrees. You can restore a session later to recreate its worktree. Active, running, needs-input, and pinned sessions are protected from cleanup.

[Agent loop] Keep track of background shells: When the agent runs a command in the background, it keeps working on other steps while the command runs. In Copilot harness sessions, a Background Shells pill above the chat input lists the shells that are still running, in both the Agents window and the editor window. Select the pill to see each shell's command and how long it has been running, and select a shell to watch its output stream as the command runs, if output is available. A shell leaves the list when its command finishes.

[Agent loop] More control for agent-triggered messages: When agents delegate work across conversations, they can adjust the delegated task as requirements change instead of waiting for the other session to finish. The send_message tool supports these actions between sessions and nested sessions on the same agent host: Steer an active conversation with a correction. Queue a follow-up as a separate turn. Replace a queued message while preserving its place in the queue. Cancel a queued message before processing begins. Queued messages run in order. After a message starts processing, it can no longer be replaced or cancelled.

[Agent UX] Arrange sessions in a grid: Compare results, monitor long-running tasks, or work across related conversations without repeatedly switching between sessions. Like in the editor window, the two-dimensional session grid in the Agents window lets you drag sessions into horizontal or vertical splits and resize the panes. Maximize a session when you need to focus, then return to the grid to continue working across conversations.

[Agent UX] Simplified new-session controls (Experimental): Settings: sessions.chat.experimental.newSessionComposerLayout , sessions.chat.unifiedWorkspacePicker.enabled (Agents window only) The experimental new-session composer keeps your prompt at the center while making session choices easier to review. Workspace, repository, and harness controls now appear in a collapsible Session Options tray above the chat input. Getting-started tips now appear in a rounded notice below the chat input, so they no longer separate the session controls from your prompt. Collapse the tray to reduce visual clutter without changing your selections. The composer remembers your choice, adapts labels as the window narrows, and keeps the options expanded when screen reader optimized mode is active. When you use a custom chat background, the tray has an opaque surface to keep its controls readable.

[Agent environments] Continue local external Copilot sessions without reloading: New conversations from the Copilot CLI and GitHub Copilot app are now picked up automatically as external sessions as soon as you send the first request, even when the Agents window is already open. You can open and continue these conversations directly from the Agents window. Use the Created In and Created Externally filters to choose which conversations appear in the sessions list.

[Agent environments] Codex chat handoff between apps: Pick up a Codex chat started in the ChatGPT app or Codex CLI and continue it in VS Code's Agents window on the same machine, with your conversation history intact. You can switch back to the ChatGPT app or Codex CLI later and continue the same chat. To use your GitHub Copilot subscription in VS Code, choose a Copilot model in the model picker. For example, you can keep working with Copilot after reaching your ChatGPT usage limit. With Codex enabled and Created Externally sessions visible, chats created or updated in ChatGPT or Codex CLI appear within a few seconds without reloading VS Code. Only one application can send messages to the chat at a time. If ChatGPT or Codex CLI still has the chat open, a This chat is open in another app banner in VS Code explains why sending is unavailable. Fully quit ChatGPT or exit the Codex CLI session, then select Retr

[Agent environments] Download remote files from the editor: Save an open remote file locally, without having to navigate for it in the Files panel. In the Agents window, open the editor's More Actions menu and select Download... to download the file to a local destination.

[Agent environments] Dev Container samples (Experimental): Setting: chat.agentHost.devContainer.samples.enabled (Agents window only) Try an agent with a ready-to-use development environment without first cloning a repository or installing its language tools locally. The Agents window now offers Dev Container samples for Go, .NET, Node.js, PHP, Python, and Rust. Install and start Docker, and then enable the setting. Open a new session in the desktop Agents window, select Dev Container Sample from the workspace picker, and send a prompt. This also requires chat.agentHost.devContainer.enabled and chat.remoteAgentHostsEnabled to be enabled.

[Chat experience] Persistent progress in chat: Settings: chat.experimental.persistentProgress , chat.experimental.persistentProgressVerbosity Persistent progress is rolling out to all users. It keeps in-progress responses visible and lets you control how much progress detail appears. Use chat.experimental.persistentProgress to choose among three icon variants, including a monochrome icon or no icon.

[MCP] See discovered MCP servers in Customizations editor (Preview): Find MCP servers supplied by plugins, extensions, and built-in integrations alongside your configured servers in the MCP Servers section of the Agent Customizations editor. Servers discovered outside your workspace's mcp.json are visible in the UI, so you don't have to use MCP: List Servers to find them. Run Chat: Open Customizations from the Command Palette and select MCP Servers to inspect the available integrations. Learn more about adding and managing MCP servers.

[Agent UX] Compact layout density: Setting: window.density.layout Fit more content in the Agents window with the Compact layout density. Compact density uses a shorter title bar, removes the gaps between workbench parts, and reduces pane spacing. By default, the Agents window follows the user-level window.density.layout setting, which also applies to editor windows. To set a different density for the Agents window only, open it and choose View > Layout Density. The density of editor windows remains unchanged.

[Agent UX] Additional session-filtering options: Focus your sessions list on the work you care about. Open Filter Sessions to filter independently by: Environment: Where the session runs, such as locally, in the cloud, or on a remote host. Harness: Which agent harness the session uses. Created In: Which application created the session, such as VS Code, Copilot CLI, or the Copilot app. The menu also brings together ordering, grouping, Show Done, and Created Externally controls.

[Agent UX] Track activity across nested sessions: Nested sessions let you split related work across separate conversations. In the Agents window, each nested session shows its unread status and last-modified time, helping you find recent activity and conversations that need your attention without opening them individually. Status is tracked independently for each conversation: Opening a nested session marks only that conversation as read. An unread nested session does not make the main session appear unread. The last-modified time reflects activity in that conversation, not activity in the main session or another nested session. Read status and last-modified times persist when you reload the window or restart VS Code. To clear unread status for an entire session tree, right-click the main session and select Mark as Read. Unrelated sessions are not affected.

[Agent UX] Improved context picker positioning: The Add Context picker now opens next to the button that invoked it and closes when you select the button again. This keeps the picker visually connected to the chat input in the editor and Agents windows.

[Agent UX] Move agent picker in context control (Experimental): Setting: sessions.chat.experimental.agentsPickerInAttachContextMenu (Agents window only) If you prefer to have the Agent picker initially hidden from the chat input to reduce the number of controls, you can configure this setting to move it into the Add Context menu. After you select an agent, the picker returns to the chat input.

[Agent UX] Customize new-session welcome messages (Experimental): Settings: sessions.chat.experimental.welcomePhrases , sessions.chat.experimental.welcomeMessages (Agents window only) New-session welcome messages add personality to the Agents window. In this release, you can add your own phrases to the defaults or replace the default phrases. Use the {name} placeholder to position your welcome name anywhere in a phrase. Select Customize Welcome Message beside the heading to change your name or open the phrase setting. You can also right-click the heading and select Hide Welcome Message to turn welcome phrases off.

[Chat experience] Meet Blobby: The /vscode-pet has a name: Blobby. Type /vscode-pet in any chat in the editor or Agents window to meet Blobby. Type /blobby to customize Blobby. More customization options become available as you explore chat and unlock achievements. Learn more about Blobby and chat pets.

[Editor experience] Frosted glass overlays and menus: Settings: workbench.modernUIFrostedGlass , workbench.modernUIFrostedGlassOpacity Menus, the Command Palette, and other pop-ups now use softly blurred, frosted glass backgrounds in the desktop Agents window. Frosted glass is also available in editor windows when the modernized UI is enabled with workbench.experimental.modernUI . Use workbench.modernUIFrostedGlassOpacity to control how much of the blurred content shows through. Lower values make the background more transparent, while higher values make it more solid. Frosted glass respects reduced-transparency preferences in VS Code and your operating system, and uses solid backgrounds with high contrast themes or when the effect isn't supported. Disable workbench.modernUIFrostedGlass if you notice display or performance issues.

[Editor experience] Choose your tab style (Experimental): Setting: workbench.experimental.modernUIEditorTabStyle When the modernized UI is enabled ( workbench.experimental.modernUI ), you can choose between two tab styles: connected: visually connect the active tab to its content. pill: display tabs as separate rounded pills.

[Editor experience] Spreading block pasting: Setting: editor.multiCursorPaste Paste a copied block at a single cursor to distribute its rows across successive destination lines. This behavior is enabled by default when editor.multiCursorPaste is set to spread. Set it to full to paste the complete block instead.

[Enterprise] Attribute Agent Host sessions to users and machines: Administrators can more easily attribute Agent Host sessions to users and machines by including operating system usernames and hostnames in VS Code's Agent Host telemetry. Merge this example into your unified managed settings, alongside your existing telemetry exporter configuration: For example, captured identity attributes can include (illustrative values): The OS username (process.user.name) and hostname (host.name) are resource attributes. The signed-in GitHub username (user.name) appears on runtime spans when supported by the bundled Copilot runtime. Identity capture is off by default, and the managed value overrides personal preferences. It is independent of content capture. VS Code's host-side control does not govern a runtime's direct exports.

[Enterprise] Migration of legacy VS Code policies to Agent Host managed settings: The Copilot harness, powered by the Copilot SDK and running in VS Code's Agent Host, is gradually becoming the default for enterprise users who have VS Code experiments enabled. To preserve existing enterprise controls during this transition, VS Code translates supported legacy policies into unified managed settings at runtime when possible. This translation applies to sessions on the same machine and does not change your policy configuration. Enterprises should still migrate to unified managed settings where applicable for consistent policy management across VS Code, the GitHub Copilot app, and Copilot CLI.

[Enterprise] Use managed settings to require sandboxing (Preview): To require sandboxing in Copilot Agent Host and prevent bypass, use these unified managed settings: Legacy VS Code sandbox settings and device policies are deprecated. ChatAgentSandboxEnabled now supplies an overridable default rather than a mandatory requirement in Agent Host. Local behavior is unchanged.

[Enterprise] Keep your selected agent through policy checks: When your enterprise requires a fresh policy check with forceRemoteSettingsRefresh, chat preserves your selected agent during policy checks instead of switching to Ask or Edit. Routine background refreshes also keep the last accepted policy active while loading. Startup, account changes, and failed refreshes continue to block submissions until policy checks succeed. Newly accepted restrictions continue to apply.

[Deprecated features and settings] Legacy VS Code sandbox settings and device policies: Legacy VS Code sandbox settings and device policies are deprecated. ChatAgentSandboxEnabled now supplies an overridable default rather than a mandatory requirement in Agent Host. Local behavior is unchanged.