Visual Studio Code 1.380, the latest update to Microsoft’s open-source code editor, introduces three new features for AI-powered coding: agent sessions in Dev Containers, an expanded Codex harness, and automated cleanup for agent sessions. Session cleanup is a preview feature.
With VS Code 1.380, released September 16, agent sessions now can be run inside a local folder’s Dev Container, where the agent uses the environment and dependencies configured for the project instead of those on the local machine. When the chat.agentHost.devContainer setting is enabled, local folders with a supported Dev Container configuration automatically show a folder menu with a “Use Dev Container” action, Microsoft said. Dev Containers require Docker to be installed on the machine.
VS Code 1.380 also expands Codex support in the agent host. This means users can continue the same Codex session between the ChatGPT app and VS Code instead of starting a new conversation, and they can switch between Copilot-backed and ChatGPT-backed models from VS Code’s model picker without losing the current conversation. If the ChatGPT app is installed and configured for computer use, then the Codex harness in VS Code can reuse that setup to interact with apps on your computer. And Codex can use the full set of tools provided by VS Code including extensions and Model Context Protocol (MCP) tools, according to Microsoft.
And VS Code 1.380 introduces a preview of automatic agent session cleanup. This feature allows developers to get finished work out of the way while preserving the conversation for later, Microsoft said. When all of an inactive session’s pull requests have merged, the Agents window can suggest marking the session as done, and can automatically delete merged sessions after a specified number of days.
Finally, with VS Code 1.380, agent automations are automatically enabled by default. They can also be exported and imported across environments or shared with other users, Microsoft said. Automations run agent tasks from a saved prompt and session configuration, either on demand or on a schedule.