OMA - One Man Army

Autonomous AI agent system for business research and strategy.

Setup

1. Create a virtual environment:

python -m venv venv
source venv/bin/activate

2. Install dependencies:

pip install -r requirements.txt

3. Configure API keys:

cp .env.example .env
# Edit .env and add your keys

Required keys:

• ANTHROPIC_API_KEY - Get from console.anthropic.com
• TAVILY_API_KEY - Get from tavily.com (free tier available)

Usage

Interactive mode:

python main.py

Single goal:

python main.py "Research the AI coding assistant market"

Architecture

OMA/
├── oma/
│   ├── core/
│   │   ├── orchestrator.py  # Central coordinator
│   │   ├── agent.py         # Base agent class
│   │   └── memory.py        # Persistent storage
│   ├── agents/
│   │   ├── researcher.py    # Web research
│   │   ├── analyst.py       # Market analysis
│   │   └── strategist.py    # Business strategy
│   └── tools/
│       ├── web.py           # Search & fetch
│       └── documents.py     # Report generation
├── data/
│   ├── memory/              # Agent learnings
│   └── outputs/             # Generated reports
└── main.py                  # Entry point

How It Works

1. Give OMA a research goal
2. Orchestrator breaks it into tasks
3. You approve the plan (checkpoint)
4. Specialist agents execute tasks
5. Results synthesized into a report