Agency Agents

Complete AI Agency - 51+ Specialized Agent Roles

Agency Agents is a comprehensive multi-agent system that simulates a complete AI agency with 51+ specialized roles covering every business function. From CEO to support staff, each agent has specific responsibilities, enabling full-service business process automation through coordinated AI collaboration. GitHub: https://github.com/msitarzewski/agency-agents Developer: msitarzewski (Community Open Source) License: MIT Key Innovation: Agency Agents creates a complete virtual company structure with specialized AI agents for every business role, enabling end-to-end business process automation with realistic organizational dynamics.

Core Philosophy: "A complete business needs complete AI coverage - every role, every function"

Why Agency Agents?

51+ Roles: Comprehensive coverage of business functions.

Real Structure: Mirrors actual company organization.

Ready Templates: Pre-built workflows for common scenarios.

Scalable: Start small, grow to full agency.

Agent Roles

Executive Layer

1. CEO - Strategic direction, final decisions
2. COO - Operations management
3. CFO - Financial oversight
4. CTO - Technology strategy
5. CMO - Marketing direction

Management Layer

6. Project Manager - Project coordination
7. Team Lead - Team management
8. Product Manager - Product strategy
9. Operations Manager - Daily operations
10. Quality Manager - Quality assurance

Specialized Roles (Sample)

11. Researcher - Information gathering
12. Analyst - Data analysis
13. Writer - Content creation
14. Editor - Content review
15. Designer - Visual design
16. Developer - Code implementation
17. Tester - Quality testing
18. DevOps - Infrastructure
19. Sales - Customer acquisition
20. Support - Customer service

... and 31+ more specialized roles

Key Features

$#1.
Role Specialization

• 51+ defined roles
• Role-specific prompts
• Specialized capabilities
• Clear responsibilities
• Expert-level performance

$#2.
Organizational Structure

• Executive layer
• Management layer
• Operational layer
• Support layer
• Clear reporting lines

$#3.
Inter-Department Collaboration

• Department communication
• Cross-functional teams
• Handoff protocols
• Escalation paths
• Meeting simulation

$#4.
Workflow Templates

• Business processes
• Project workflows
• Approval chains
• Quality gates
• Reporting cycles

$#5.
Management Features

• Performance tracking
• Resource allocation
• Task prioritization
• Conflict resolution
• Decision escalation

Installation

Docker

# Pull image
docker pull agencyagents/agency:latest

# Run container
docker run -d \
  --name agency \
  -p 8080:8080 \
  -v ~/agency/data:/app/data \
  agencyagents/agency:latest

Manual Installation

# Clone repository
git clone https://github.com/msitarzewski/agency-agents.git
cd agency-agents

# Install dependencies
pip install -r requirements.txt
npm install

# Configure
cp config.example.yaml config.yaml

# Run
python main.py

Quick Start

# Install package
pip install agency-agents

# Initialize
agency-agents init my-agency
cd my-agency

# Start agency
agency-agents start

Configuration

config.yaml

agency:
  name: "My AI Agency"
  size: "full"  # small, medium, full
  
  departments:
    executive: { enabled: true }
    management: { enabled: true }
    research: { enabled: true }
    content: { enabled: true }
    development: { enabled: true }
    operations: { enabled: true }
    support: { enabled: true }

ai:
  default_model: "gpt-4o"
  fallback_model: "gpt-4o-mini"
  
  role_models:
    ceo: "gpt-4o"
    researcher: "gpt-4o"
    writer: "claude-sonnet-4"
    developer: "gpt-4o"

server:
  port: 8080
  host: "0.0.0.0"

Usage

Submit a Project

curl -X POST http://localhost:8080/api/projects \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Market Research Report",
    "department": "research",
    "priority": "high",
    "deadline": "2026-04-01"
  }'

Agency Dashboard

Access: http://localhost:8080/dashboard
View organizational chart
Monitor agent activities
Track project progress
Review outputs

Role Commands

/assign researcher "Research AI market trends"
/assign writer "Draft report based on research"
/assign editor "Review and edit the report"
/status - Show all active projects

Pricing

Free: Agency Agents is completely free and open source under MIT license. You only pay for the AI models you use through their respective providers.

System Requirements

Use Cases

Business Process Automation

Automate complete business processes with specialized agents.

Virtual AI Company

Run a fully AI-staffed virtual company.

Multi-Department Workflows

Execute workflows spanning multiple departments.

Enterprise Simulation

Simulate enterprise operations for planning and testing.

Educational Demonstrations

Teach business processes and AI coordination.

Research on Agent Organizations

Study emergent behaviors in agent organizations.

Community and Support

• GitHub: https://github.com/msitarzewski/agency-agents
• Documentation: https://github.com/msitarzewski/agency-agents#readme
• Issues: GitHub Issues tab
• Discussions: GitHub Discussions tab

FAQ

Is Agency Agents free to use?

Yes, Agency Agents is free and open source (MIT license). You only pay for AI model API costs if using external models.

What are the system requirements for Agency Agents?

Agency Agents requires 4GB RAM of RAM minimum. Runtime: Python/Node.js. It runs on Windows, macOS, and Linux.

Can I self-host Agency Agents?

Yes. Agency Agents is open source (MIT) and can be self-hosted on your own hardware. Clone the repository from GitHub and follow the installation guide.

How does Agency Agents compare to OpenClaw?

Agency Agents offers a different approach compared to OpenClaw. While OpenClaw provides the largest ecosystem with 13,729+ skills and maximum flexibility, Agency Agents focuses on multi-agent. Choose Agency Agents if you prioritize its specific features; choose OpenClaw for the broadest compatibility and community support.

Is Agency Agents suitable for beginners?

Agency Agents requires some technical knowledge to set up (Python/Node.js). If you are a beginner, consider starting with QClaw (one-click install) or MaxClaw (cloud-based, no setup) first, then graduate to Agency Agents as you gain experience.

License

MIT License - Free for personal and commercial use.

Tags

multi-agent, agency, business, specialized-roles, 51-agents, enterprise, organization, automation