🤖 Claude Agents Manager

Elite AI Research & Development Platform with 60+ Specialized Agents

🚀 Quick Installation

npm install -g claude-agents-manager

Start using 60+ specialized AI agents in seconds!

⚡ 5-Minute Quick Wins

🔍 Instant Code Review

Get enterprise-grade code analysis in seconds

claude-agents run architect-review --task "Review my React components for best practices"

📊 Elite Market Research

Generate citation-backed research reports

claude-agents research "AI automation trends 2024" --depth comprehensive

🏗️ Instant API Documentation

Create professional OpenAPI specs

claude-agents run api-documenter --task "Document my REST API endpoints"

🎯 Strategic Project Planning

Break down complex projects into actionable tasks

claude-agents run project-planner --task "Plan microservices migration"

📚 Complete Commands Reference

Command Description Example
🚀 Setup & Installation
claude-agents init
Initialize agents in current project with context-forge awareness Sets up .claude directory and agent commands
claude-agents install
Interactive agent installation with category selection Choose from 60+ specialized agents
claude-agents install --all
Install all available agents at once Full agent suite installation
claude-agents uninstall --all --clean
Remove all agents and clean up directories Complete cleanup with directory removal
🤖 Agent Management
claude-agents list
Display all available agents organized by categories Browse 60+ agents across 12 specializations
claude-agents list --installed
Show only currently installed agents View your active agent toolkit
claude-agents info [agent-name]
Get detailed information about a specific agent claude-agents info ai-engineer
claude-agents enable [agent]
Enable a previously disabled agent claude-agents enable code-reviewer
claude-agents disable [agent]
Temporarily disable an agent without removing it claude-agents disable test-runner
claude-agents remove [agent]
Permanently remove an installed agent claude-agents remove debugger
⚡ Agent Execution
claude-agents run [agent] --task "description"
Execute an agent with a specific task description claude-agents run frontend-developer --task "Create login component"
claude-agents run [agent] -i
Run agent in interactive mode for complex tasks claude-agents run project-planner -i
claude-agents run [agent] --file [path]
Run agent with specific file or directory context claude-agents run refactor --file src/components/
🔬 Elite Research
claude-agents research --setup
Install and configure all 12 Elite Research Team agents One-time setup for research capabilities
claude-agents research --standard "topic"
Execute standard research workflow (5 agents, ~10 min) claude-agents research --standard "AI market trends"
claude-agents research --deep "topic"
Execute deep research workflow (12 agents, ~25 min) claude-agents research --deep "blockchain scalability"
claude-agents research
Interactive research management and coordination GUI for custom research workflows
🔄 Workflow Orchestration
claude-agents workflow --list
Display available workflow templates Browse pre-built enterprise workflows
claude-agents workflow --template [name]
Execute a specific workflow template claude-agents workflow --template fullstack-development
claude-agents workflow --status [id]
Check execution status of a running workflow Monitor multi-agent coordination progress
claude-agents workflow
Interactive workflow designer and executor Create custom multi-agent workflows
🎨 Customization & Management
claude-agents customize
Interactive enterprise customization suite Enhance agents with company-specific rules
claude-agents customize --agent [name]
Customize a specific agent with templates claude-agents customize --agent api-developer
claude-agents create
Create a new custom agent interactively Build agents for specialized workflows
claude-agents dashboard
Launch web dashboard on port 7842 GUI for agent management and monitoring
claude-agents dashboard --port [port]
Launch dashboard on custom port claude-agents dashboard --port 8080
🛠️ Development & Testing
npm test
Run all test suites including memory and concurrent execution Comprehensive test coverage validation
npm run lint
Run ESLint on all source files with CLI-optimized rules Code quality and style validation
npm run demo
Execute enhanced workflow demonstration See agents in action with sample tasks
npm run research-demo
Execute research workflow example Demo of Elite Research Team coordination
❓ Help & Information
claude-agents --help
Display comprehensive help information Full command reference and usage
claude-agents --version
Show current version of claude-agents-manager Version information for troubleshooting

🎯 Essential Commands

Basic Operations

# List all available agents claude-agents list # Get agent information claude-agents info ai-engineer # Run an agent with a task claude-agents run [agent] --task "description" # Interactive agent execution claude-agents run [agent] -i

Research Commands

# Standard research (5 agents) claude-agents research "topic" --type standard # Deep research (12 agents) claude-agents research "topic" --type deep # Custom research workflow claude-agents research "topic" --agents "expert-synthesizer,fact-checker"

Workflow Management

# List workflow templates claude-agents workflow --list # Execute pre-built workflow claude-agents workflow --template fullstack-development # Create custom workflow claude-agents workflow

🏢 60+ Specialized Agents by Category

🏗️ Software Architecture (4)
architect-review - Architecture validation
backend-architect - Backend design
graphql-architect - GraphQL systems
project-planner - Strategic planning
💻 Software Engineering (5)
frontend-developer - Modern web apps
mobile-developer - iOS/Android/React Native
refactor - Code improvement
tdd-specialist - Test-driven development
test-automator - Automated testing
🧠 AI & Machine Learning (4)
ai-engineer - AI system development
ml-engineer - ML pipeline engineering
mlops-engineer - ML operations
prompt-engineer - Prompt optimization
🔬 Elite Research (12)
research-coordinator - Research orchestration
deep-miner - Academic & expert networks
expert-synthesizer - Expert consensus
fact-checker - Source verification
executive-summarizer - C-suite summaries
report-generator - Professional reports
☁️ Cloud Infrastructure (4)
cloud-architect - Multi-cloud design
deployment-engineer - CI/CD pipelines
devops-troubleshooter - System diagnostics
terraform-specialist - Infrastructure as code
💼 Business Strategy (3)
business-analyst - Business analysis
marketing-writer - Technical marketing
product-manager - Product strategy
📊 Data Engineering (3)
data-engineer - Data pipelines
data-scientist - Statistical analysis
quant-analyst - Quantitative analysis
🔒 Security Operations (3)
security-auditor - Security assessment
incident-responder - Security incidents
risk-manager - Risk assessment

🎮 How to Use Software Agents

1. 🎯 Choose Your Agent

Select the right specialist for your task:

# Browse available agents claude-agents list # Get detailed agent information claude-agents info frontend-developer

2. 🗣️ Define Your Task

Be specific about what you want the agent to accomplish:

# Good: Specific and actionable claude-agents run frontend-developer --task "Create a responsive React component for user authentication with form validation" # Better: Include context and constraints claude-agents run backend-architect --task "Design a microservices architecture for e-commerce platform handling 10k concurrent users with PostgreSQL"

3. 🚀 Execute with Options

Use different execution modes based on your needs:

# Standard execution claude-agents run ai-engineer --task "Optimize ML model performance" # Interactive mode for complex tasks claude-agents run project-planner -i # With file context claude-agents run refactor --task "Improve code quality" --file src/components/UserForm.js

4. 🔄 Coordinate Multiple Agents

Use workflows for complex multi-step processes:

# Use pre-built workflows claude-agents workflow --template fullstack-development # Create custom multi-agent workflow claude-agents workflow

5. 📊 Monitor & Track Progress

Keep track of agent activities and results:

# View agent execution history claude-agents workflow --status [workflow-id] # Check memory for coordination claude-agents info --memory

🔬 Elite Research Workflows

Transform complex research challenges into comprehensive, citation-backed reports

📋 Standard Research (5 agents, ~10 minutes)

claude-agents research "blockchain scalability solutions" --type standard

Perfect for: Market overviews, competitive analysis, technology assessments

🎯 Deep Research (12 agents, ~25 minutes)

claude-agents research "enterprise AI adoption strategies" --type deep

Perfect for: Strategic decisions, investment research, comprehensive analysis

🔧 Custom Research Workflows

claude-agents research "quantum computing applications" --agents "expert-synthesizer,fact-checker,trend-analyst"

Perfect for: Specialized investigations, targeted analysis, custom requirements

💡 Pro Tips & Best Practices

🎯 Task Clarity

Be specific about deliverables, constraints, and success criteria. Include context like tech stack, team size, or business requirements.

🔄 Agent Coordination

Use the shared memory system to coordinate multiple agents. Results from one agent automatically inform others in workflows.

📊 Research Quality

For research tasks, specify the depth needed (overview vs deep-dive) and any particular perspectives or industries to focus on.

🏗️ Architecture Reviews

Include system scale, performance requirements, and security considerations when requesting architecture reviews.

⚡ Concurrent Execution

The system automatically runs related operations in parallel. Design workflows to take advantage of this for maximum efficiency.

🔍 Context-Forge Integration

If you're using context-forge projects, agents automatically respect existing PRPs, implementation plans, and project rules.

🚀 Advanced Features

Enterprise Customization

# Interactive customization suite claude-agents customize # Bulk enhance agents claude-agents customize --bulk-enhance

Web Dashboard

# Launch management dashboard claude-agents dashboard # Access at http://localhost:7842

Agent Management

# Install specific agents claude-agents install architect-review # Create custom agents claude-agents create # Remove agents claude-agents remove [agent-name]

⚡ Performance Features

🔄 Concurrent Execution

All operations run in parallel for maximum speed

🧠 Shared Memory System

Agents coordinate seamlessly through persistent memory

📚 Context-Forge Integration

Respects existing project structure and conventions

🎯 Enterprise Quality

Professional-grade outputs with citations and validation

📞 Need Help?

Run claude-agents --help for detailed command information

Visit our GitHub repository for documentation and examples

claude-agents info [agent-name] # Get detailed agent information claude-agents list --category research # Filter agents by category