Skip to content

01-getting-started.md

Latest commit

 

History

History
270 lines (205 loc) · 5.92 KB

01-getting-started.md

File metadata and controls

270 lines (205 loc) · 5.92 KB

Getting Started with Claude-Flow

Welcome to Claude-Flow, an advanced AI agent orchestration system designed for sophisticated multi-agent collaboration, task coordination, and memory management. This guide will help you get up and running quickly.

Quick Installation

Option 1: NPX (Recommended for First-Time Users)

# Run directly without installation
npx claude-flow

# Or install globally for persistent use
npm install -g claude-flow
claude-flow --version

Option 2: Deno Installation

# Install with Deno
deno install --allow-all --name claude-flow https://raw.githubusercontent.com/ruvnet/claude-code-flow/main/src/cli/index.ts

# Verify installation
claude-flow --help

Option 3: From Source

# Clone the repository
git clone https://github.com/ruvnet/claude-code-flow.git
cd claude-code-flow

# Install dependencies and build
deno task install

# Run from source
deno task dev

Initial Setup

1. Initialize Configuration

# Create default configuration file
claude-flow config init

# Verify configuration
claude-flow config show

2. Start the Orchestrator

# Basic start
claude-flow start

# Start with daemon mode for background operation
claude-flow start --daemon

# Start with custom port
claude-flow start --port 3000

3. Verify System Health

# Check system status
claude-flow agent list
claude-flow memory stats
claude-flow mcp status

Your First Workflow

Let's create a simple research workflow to demonstrate Claude-Flow's capabilities:

Step 1: Spawn a Research Agent

# Create a research agent
claude-flow agent spawn researcher --name "Research Assistant"

# Verify agent is active
claude-flow agent list

Step 2: Create a Research Task

# Create a research task
claude-flow task create research "Analyze current trends in AI development tools" \
  --priority high \
  --estimated-duration 2h

# Check task status
claude-flow task list

Step 3: Monitor Progress

# Monitor task execution
claude-flow task monitor --follow

# Check agent activity
claude-flow agent info <agent-id>

Step 4: Review Results

# Query memory for research findings
claude-flow memory query --filter "AI development tools" --recent

# Export findings
claude-flow memory export --filter "research-results" --output research-findings.json

Interactive Exploration

For learning and experimentation, use the interactive REPL mode:

# Start interactive session
claude-flow repl

In REPL mode, you can:

# Get help
> help

# Spawn agents interactively
> agent spawn coordinator --name "Project Manager"

# Create tasks
> task create analysis "Evaluate system performance"

# Query memory
> memory query --recent --limit 5

# Exit REPL
> exit

Basic Concepts

Agents

Agents are specialized AI workers with specific capabilities:

  • Researcher: Information gathering and analysis
  • Implementer: Code development and technical tasks
  • Analyst: Data analysis and pattern recognition
  • Coordinator: Planning and task delegation

Tasks

Tasks represent work to be done:

  • Research: Information gathering
  • Implementation: Code development
  • Analysis: Data processing
  • Coordination: Planning and management

Memory Bank

The memory system stores:

  • Agent discoveries and insights
  • Task progress and results
  • Shared knowledge across agents
  • Project history and context

MCP Integration

Model Context Protocol enables:

  • External tool integration
  • API connectivity
  • Custom tool development
  • Secure tool execution

Common Commands Reference

Agent Management

# List all agents
claude-flow agent list

# Get detailed agent info
claude-flow agent info <agent-id>

# Terminate an agent
claude-flow agent terminate <agent-id>

Task Management

# List all tasks
claude-flow task list

# Check task status
claude-flow task status <task-id>

# Cancel a task
claude-flow task cancel <task-id>

Memory Operations

# Search memory
claude-flow memory query --search "keyword"

# View memory stats
claude-flow memory stats

# Clean up old entries
claude-flow memory cleanup --older-than 30d

Configuration

# View current config
claude-flow config show

# Update settings
claude-flow config set orchestrator.maxConcurrentAgents 15

# Reset to defaults
claude-flow config init --force

Next Steps

  1. Explore the Architecture: Read 02-architecture-overview.md to understand how Claude-Flow works
  2. Configure Your System: See 03-configuration-guide.md for detailed configuration options
  3. Learn Agent Management: Check 04-agent-management.md for advanced agent patterns
  4. Create Complex Workflows: Study 05-task-coordination.md for workflow orchestration

Getting Help

Troubleshooting Common Issues

Installation Problems

If you encounter permission issues:

# For NPM
npm config set prefix ~/.npm-global
export PATH=~/.npm-global/bin:$PATH

# For Deno
export PATH="$HOME/.deno/bin:$PATH"

Configuration Issues

If configuration fails to initialize:

# Check directory permissions
ls -la $(pwd)

# Manually create config
touch claude-flow.config.json
claude-flow config init --force

Agent Startup Issues

If agents fail to start:

# Check system resources
claude-flow system resources

# Increase limits
claude-flow config set orchestrator.maxConcurrentAgents 5

# Check logs
claude-flow logs --level debug

You're now ready to start using Claude-Flow! Continue to the next sections for more advanced features and configuration options.