Source Intelligence

DisclaimerUnofficial, and not affiliated with Anthropic. Nearly all of this is read straight out of what ships: npm bundles, captured prompts, published docs. Anthropic's own notes go in verbatim, marked as theirs. The rest is my reading, and every entry carries the strings behind it. If one looks wrong, vote it down and say why.

All of v1.0.28 Home All releases olderv1.0.27 v1.0.29newer
Claude Code v1.0.28

New Features

Model Selection Command Menu

  • Added new interactive model selection functionality that allows users to switch between different Claude models during a session
  • When invoked, users can navigate through available models and select their preferred model
  • Includes keyboard shortcuts: ESC key to cancel and keep current model
  • The selected model is displayed in bold text for clarity
  • Example usage: The menu will show available models and allow selection via arrow keys

Enhanced MCP (Model Context Protocol) Support

  • Added MAX_MCP_OUTPUT_TOKENS environment variable to control the maximum tokens for MCP output
  • Default value: 25,000 tokens (configurable via environment variable)
  • Example: export MAX_MCP_OUTPUT_TOKENS=50000 to increase the limit

Improved Resume Command

  • Refactored the resume command to use a more robust component-based architecture
  • Better error handling when loading conversations
  • Async loading with proper state management
  • Shows loading state while fetching conversations
  • Filters out sidechain conversations automatically

See this entry in the whole of v1.0.28 →