Alpha
Internal or limited testing with significant changes expected. May have bugs or incomplete functionality.
Claude Code continues to evolve rapidly with experimental features and beta programs. This guide explains how to access, test, and provide feedback on cutting-edge capabilities before they reach general availability.
Alpha
Internal or limited testing with significant changes expected. May have bugs or incomplete functionality.
Beta
Feature-complete but gathering feedback. Available to specific user groups or plans.
Preview
Near-final implementation. Broadly available but may have minor adjustments.
GA
Generally Available. Stable, supported, and recommended for production use.
Multi-agent collaboration where Claude Code spawns and coordinates teammate agents:
# Enable agent teams (research preview)export CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1Features:
TeammateIdle and TaskCompleted hooks for coordination (v2.1.33)user, project, or local scopeTask(agent_type) syntax in agent tools frontmatterFast mode uses the same Claude Opus 4.6 model with faster output — no model switch.
# Toggle fast mode inside a session/fastClaude Code GitHub Actions enables AI-powered automation in your GitHub workflow:
name: Claude Code Reviewon: [pull_request, issue_comment]
jobs: claude-action: runs-on: ubuntu-latest steps: - uses: anthropics/claude-code-action@beta with: github-token: ${{ secrets.GITHUB_TOKEN }} anthropic-api-key: ${{ secrets.ANTHROPIC_API_KEY }}Features:
@claude mentions in issues and PRsAn experimental installation method that provides platform-native binaries:
# From existing Claude Code installationclaude install
# Fresh install via scriptcurl -fsSL claude.ai/install.sh | bashDesign-to-code integration through Model Context Protocol:
# Enable in Figma Desktop App# Preferences → Dev Mode → Enable MCP Server
# Add to Claude Codeclaude mcp add --transport sse figma-dev-mode http://127.0.0.1:3845/sseCapabilities:
CLAUDE_CODE_SHELL
Override automatic shell detection (v2.0.65)
export CLAUDE_CODE_SHELL=/bin/zshCLAUDE_CODE_PROXY_RESOLVES_HOSTS
Enable proxy DNS resolution (opt-in since v2.0.55)
export CLAUDE_CODE_PROXY_RESOLVES_HOSTS=trueCLAUDE_BASH_MAINTAIN_PROJECT_WORKING_DIR
Lock Bash tool to project root directory
export CLAUDE_BASH_MAINTAIN_PROJECT_WORKING_DIR=1MCP_TIMEOUT
Adjust MCP server startup timeout (milliseconds)
export MCP_TIMEOUT=30000The MCP ecosystem includes many production-ready servers:
As of version 2.0.58, Claude Code is available to:
# View current version and featuresclaude --version
# Check system informationclaude doctor
# View available CLI flagsclaude --help
# Check MCP server status (terminal command)claude mcp listBased on recent releases, these features have graduated from beta to GA:
Claude Code for Desktop
Released in v2.0.51 - Standalone desktop application (Download)
Background Agents
Released in v2.0.60 - Run agents in background while you work
Named Sessions
Released in v2.0.64 - Use /rename and /resume <name> for session management
Rules Directory
Released in v2.0.64 - New .claude/rules/ directory for memory organization
Prompt Suggestions
Released in v2.0.67 - Tab to accept AI-suggested prompts
Model Switching
Released in v2.0.65 - Switch models with alt+p / option+p while typing
Claude Opus 4.6
Released in v2.1.32 - Latest frontier model with enhanced capabilities
Automatic Memories
Released in v2.1.32 - Claude records and recalls memories automatically
Hooks System
Released in v1.0.38 - User-defined automation points
/feedback command in Claude Code# Create test environmentmkdir claude-test && cd claude-testgit init
# Test with a fresh project directoryclaude
# Use temporary settings directoryexport CLAUDE_CONFIG_DIR=~/.claude-test// In ~/.claude/settings.json (user scope){ "autoUpdates": false}# Or use environment variable for debug loggingexport ANTHROPIC_LOG=debugBased on community discussions and development patterns:
Enhanced CI/CD
Deeper integration with build pipelines beyond GitHub Actions
Team Collaboration
Shared contexts and collaborative coding sessions
Plugin System
Extensible architecture for custom integrations
Performance Mode
Optimized for large-scale refactoring operations
Environment Details
claude doctor > diagnostic.txtReproduction Steps
Submit via Appropriate Channel
/feedback for suggestionsVersion Control
Always commit before testing new features
Incremental Testing
Start with non-critical projects
Document Findings
Keep notes on behavior changes
Share Knowledge
Help others with your discoveries
MCP Server Connection Problems:
# Debug MCP connectionsclaude --debug "mcp"
# List configured MCP serversclaude mcp list
# To fix a broken server, remove and re-add itclaude mcp remove <server-name>claude mcp add <server-name> <command> [args...]Permission Issues:
# View and manage permissions interactively/permissions# Use the interactive prompt to allow or deny specific toolsModel Access:
# Check available models (inside a REPL session)/model
# Upgrade plan if needed (inside a REPL session)/upgrade