Experimental feature allows users to get a second opinion from a complementary model on the primary agent’s work.

VS Code 1.135 brings GitHub Copilot's cross-model critic into Agent Host sessions, while also adding external session continuation, Agents window refinements and more detailed token-usage reporting.

Multi-chat Claude sessions allow VS Code users to run multiple related conversation threads in one Claude agent session instead of separate top-level sessions. Visual Studio Code 1.128, the most ...

Microsoft is bringing five more Claude-native capabilities to Azure-hosted deployments, extending a pattern of accommodating Anthropic's models, APIs and agent tooling across Microsoft and GitHub ...

Anthropic's Claude Code for VS Code has surpassed OpenAI's Codex extension in Visual Studio Code Marketplace installs and review volume despite launching later. OpenAI positions Codex as an IDE ...