Skip to content

Getting Started

Jump to bottom
Prakash Tiwari edited this page Aug 11, 2025 · 1 revision

Getting Started with OSA

This guide will help you get OSA up and running in minutes.

📋 Prerequisites

  • Python: 3.9 or higher
  • Memory: 8GB RAM minimum (16GB recommended)
  • Disk Space: 2GB free space
  • OS: Linux, macOS, or Windows

🚀 Installation

1. Clone the Repository

git clone https://github.com/prakashgbid/omnimind.git
cd omnimind

2. Create Virtual Environment

# Create virtual environment
python -m venv venv

# Activate it
# On Linux/macOS:
source venv/bin/activate

# On Windows:
venv\Scripts\activate

3. Install Dependencies

pip install -r requirements.txt

4. (Optional) Install Ollama for Local LLMs

# Linux/macOS
curl -fsSL https://ollama.ai/install.sh | sh

# Pull models
ollama pull llama3.2:3b
ollama pull mistral
ollama pull deepseek-coder

⚙️ Configuration

Environment Variables

Create a .env file in the project root:

# Optional: API Keys for cloud providers
OPENAI_API_KEY=your_key_here
ANTHROPIC_API_KEY=your_key_here

# Optional: Custom settings
OSA_MAX_INSTANCES=10
OSA_THINKING_DEPTH=10
OSA_LEARNING_ENABLED=true

Local Configuration

OSA works out of the box with local models. No API keys required!

🎮 Running OSA

Interactive Mode

The easiest way to start:

python run_complete_osa.py

You'll see a menu with options:

  1. Deep thinking & task accomplishment
  2. Leadership & delegation
  3. Problem solving with alternatives
  4. Continuous thinking demonstration
  5. Complex goal breakdown
  6. Blocker handling
  7. Show thinking status

Python Script

import asyncio
from osa_complete_final import create_complete_osa

async def main():
    # Create OSA instance
    osa = await create_complete_osa()
    
    # Give it a task
    result = await osa.think_and_accomplish(
        "Create a todo list app with user authentication"
    )
    
    print(f"Task completed: {result['summary']}")
    print(f"Thoughts generated: {result['thinking_insights']['total_thoughts']}")

asyncio.run(main())

Web Monitoring

Start the real-time monitoring interface:

# Terminal 1: Start WebSocket server
python src/osa_logger.py

# Terminal 2: Run OSA
python run_complete_osa.py

# Open in browser
open web/index.html

🧠 Basic Usage Examples

Simple Task

# Initialize
osa = await create_complete_osa()

# Simple task
result = await osa.accomplish("Write a Python function to sort a list")

Complex Project

# Lead a project
project = await osa.lead_complex_project(
    project_name="E-Commerce Platform",
    requirements=[
        "User authentication",
        "Product catalog",
        "Shopping cart",
        "Payment processing"
    ],
    team_size=5
)

Problem Solving

# Solve with alternatives
solution = await osa.solve_with_alternatives(
    "Fix database performance issues when indexes are missing"
)

print(f"Solutions found: {solution['alternatives_available']}")

Continuous Thinking

# Let OSA think about a topic
thoughts = await osa.think_continuously_about(
    topic="How to make apps go viral",
    duration_seconds=30
)

print(f"Generated {thoughts['thoughts_generated']} thoughts")

📊 Understanding the Output

OSA provides detailed insights into its thinking:

result = await osa.think_and_accomplish("Build a chat app")

# Access thinking insights
print(f"Total thoughts: {result['thinking_insights']['total_thoughts']}")
print(f"Reasoning chains: {result['thinking_insights']['reasoning_chains']}")
print(f"Blockers handled: {result['thinking_insights']['blockers_handled']}")
print(f"Alternatives available: {result['thinking_insights']['alternatives_available']}")
print(f"Confidence: {result['thinking_insights']['confidence']:.1%}")

# View thought graph
print(result['thinking_insights']['thinking_visualization'])

🔧 Troubleshooting

Common Issues

Python version error

# Check Python version
python --version

# Use python3 if needed
python3 run_complete_osa.py

Memory issues

# Reduce parallel instances
osa = await create_complete_osa(max_claude_instances=3)

Import errors

# Ensure you're in the project directory
cd omnimind

# Reinstall dependencies
pip install -r requirements.txt

📚 Next Steps

💡 Tips

  1. Start small: Begin with simple tasks to understand OSA's capabilities
  2. Watch the monitor: Use the web interface to see OSA's thinking in real-time
  3. Learn from patterns: OSA gets smarter with each task
  4. Experiment: Try different types of problems to see OSA's versatility

Need help? Open an issue or join our discussions!

Clone this wiki locally