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 v0.2.31 Home All releases olderv0.2.30 v0.2.32newer
Claude Code v0.2.31

New Features

Custom Slash Commands

Users can now create custom slash commands by placing Markdown files in .claude/commands/ directories. These files will automatically appear as slash commands that insert prompts into your conversation.

Usage Example:

# Create a custom command file
mkdir -p .claude/commands/
echo "# My Custom Prompt\nThis is my custom prompt content" > .claude/commands/my-command.md

# In Claude Code, use the command
/my-command
# This will insert the contents of my-command.md into your conversation

This feature allows you to create reusable prompts and workflows specific to your project or personal preferences.

MCP Debug Mode

A new --mcp-debug flag provides detailed error information when MCP (Model Context Protocol) servers encounter issues.

Usage Example:

# Run Claude Code with MCP debugging enabled
claude --mcp-debug

# When MCP server errors occur, you'll see detailed error messages like:
# MCP server "server-name" Error details here

This helps troubleshoot MCP server configuration and connectivity issues.

See this entry in the whole of v0.2.31 →