Quick Start

⚡ Quick Start Guide

Get up and running with Claude-Flow in under 5 minutes!

🚀 Installation

If you haven't installed Claude-Flow yet, follow these quick steps:

# Install Claude Code (required)
npm install -g @anthropic-ai/claude-code
claude --dangerously-skip-permissions

# Install Claude-Flow Alpha
npm install -g claude-flow@alpha

# Verify installation
claude-flow --version

🐝 Your First Hive-Mind Swarm

1. Initialize a Simple Swarm

# Create a mesh topology with 3 agents
claude-flow hive init --topology mesh --agents 3

# Check swarm status
claude-flow hive status

Expected Output:

🐝 Hive-Mind Initialized
├── Topology: mesh
├── Agents: 3/3 active
├── Memory: SQLite initialized
└── Status: Ready for orchestration

2. Your First Orchestrated Task

# Create a simple hello world project
claude-flow orchestrate "create a hello world API with tests" --agents 3 --parallel

This command will:

• ✅ Spawn 3 specialized agents (architect, coder, tester)
• ✅ Create project structure automatically
• ✅ Generate API code with Express.js
• ✅ Write comprehensive tests
• ✅ Set up package.json with dependencies

3. Monitor Real-Time Progress

# Watch swarm coordination in real-time
claude-flow hive monitor --live

🧪 SPARC Development Mode

Test-Driven Development with AI

# Run SPARC TDD workflow
claude-flow sparc run tdd "user authentication system"

This will execute the complete SPARC methodology:

1. Specification: Analyze requirements
2. Pseudocode: Design algorithms
3. Architecture: Structure components
4. Refinement: Implement with TDD
5. Completion: Integration and testing

Available SPARC Modes

# List all SPARC modes
claude-flow sparc modes

# Run specific development patterns
claude-flow sparc run dev "build REST API"        # Development mode
claude-flow sparc run api "user management API"   # API-focused mode
claude-flow sparc run ui "dashboard component"    # UI development mode
claude-flow sparc run test "integration tests"    # Testing mode
claude-flow sparc run refactor "optimize database queries"  # Refactoring mode

🔧 MCP Tools in Action

Explore Available Tools

# List all 87 MCP tools
claude-flow mcp tools list

# Get detailed tool information
claude-flow mcp tools info swarm_init
claude-flow mcp tools info neural_train

Execute Tools Directly

# Initialize neural patterns
claude-flow mcp execute neural_train --pattern_type coordination --training_data "development patterns"

# Store project context in memory
claude-flow mcp execute memory_usage --action store --key "project/config" --value "API project settings"

# Generate performance report
claude-flow mcp execute performance_report --format detailed --timeframe 24h

🧠 Memory System Usage

Store and Retrieve Context

# Store project decisions
claude-flow memory store "architecture/decisions" "Using microservices with Redis cache"

# Recall previous decisions
claude-flow memory recall "architecture/*"

# Search memory
claude-flow memory search "authentication" --limit 5

Cross-Session Persistence

# Export current session
claude-flow memory export --session current --format json

# Restore previous session
claude-flow memory restore --session "project-alpha-v1"

📊 GitHub Integration

Repository Analysis

# Analyze repository with AI
claude-flow github analyze --repo owner/repository --deep

# Enhance pull requests automatically
claude-flow github pr enhance --pr 123 --add-tests --improve-docs

# Automated code review
claude-flow github review --pr 123 --style comprehensive

Workflow Automation

# Create GitHub workflow
claude-flow github workflow create --name "AI Code Review" \
  --trigger pr --actions "review,test,format"

🎯 Real-World Examples

Example 1: Full-Stack Application

# Create complete full-stack app
claude-flow orchestrate \
  "build a task management app with React frontend, Node.js API, and PostgreSQL database" \
  --agents 8 --topology hierarchical --parallel

Example 2: API with Documentation

# Generate REST API with docs
claude-flow sparc run api "user management system with OpenAPI docs and Swagger UI"

Example 3: Testing Suite

# Create comprehensive test suite
claude-flow sparc run test "integration tests for authentication API with mocking"

📈 Performance Monitoring

Real-Time Metrics

# Monitor swarm performance
claude-flow metrics --live --interval 5s

# Generate performance report
claude-flow performance report --format summary

Optimization

# Analyze bottlenecks
claude-flow analyze bottlenecks --component swarm

# Optimize coordination
claude-flow optimize --target speed --metric response-time

🛠️ Configuration

Customize Behavior

# Set default topology
claude-flow config set hive.defaultTopology hierarchical

# Configure memory retention
claude-flow config set memory.retention 30d

# Enable advanced hooks
claude-flow hooks enable --post-edit --pre-task

Environment Variables

# Set environment variables
export CLAUDE_FLOW_MAX_AGENTS=12
export CLAUDE_FLOW_MEMORY_SIZE=2GB
export CLAUDE_FLOW_ENABLE_NEURAL=true

🚨 Common Quick Fixes

Reset Everything

# Reset Claude-Flow to clean state
claude-flow reset --hard

# Reinitialize
claude-flow init

Check System Health

# Comprehensive health check
claude-flow health check --verbose

# Test all integrations
claude-flow test integrations

📚 What's Next?

Now that you have Claude-Flow running, explore these advanced topics:

• Hive-Mind Intelligence - Deep dive into AI coordination
• Neural Networks - Understanding AI pattern recognition
• MCP Tools - Complete reference for all 87 tools
• SPARC Methodology - Master test-driven development
• Development Patterns - Best practices and patterns

🤝 Getting Help

• Documentation: Browse the full wiki
• Examples: Check out real-world examples
• Issues: Report bugs
• Community: Join Discord

---

🎉 Congratulations! You're now ready to build with AI-powered orchestration. Start with simple tasks and gradually explore more complex workflows as you become familiar with Claude-Flow's capabilities.