Skip to content

CraigFineTuned/OmnitrAIce

Repository files navigation

OmnitrAIce: Revolutionary Project Generation System

Created by CraigFineTuned - Building revolutionary software with small, efficient models

Purpose

OmnitrAIce demonstrates how small, efficient language models can produce exceptional results through first-principles thinking. By focusing on fundamental truths rather than model size, this system achieves high-quality outputs that rival much larger models while maintaining speed and accessibility.

Features

  • First-Principles Thinking: Break down problems to fundamental components and eliminate artificial constraints
  • Multi-Agent Collaboration: Specialized agents work together (CEO, CTO, Architect, Developer, Filesystem)
  • Revolutionary Code Generation: Generate actual implementation code using first-principles approach
  • File Structure Creation: Create complete project structures optimized for revolutionary projects
  • Interactive Web UI: Customize agents and generate projects through a user-friendly interface
  • Revolutionary Metrics: Measure innovation score, disruption factor, and 10x improvement score
  • YAML/JSON Configuration: Configure all aspects of the system through external configuration files

Modular Architecture

OmnitrAIce follows a modular architecture with clear separation of concerns:

For all available execution methods, see EXECUTION_METHODS.md. omnitrace/ ├── agents/ # Specialized agent implementations │ ├── agent_interface.py # Protocol defining agent interface │ ├── ceo_agent.py # CEO agent for vision creation │ ├── cto_agent.py # CTO agent for technical strategy │ ├── architect_agent.py # Architect agent for system design │ ├── developer_agent.py # Developer agent for implementation │ └── filesystem_agent.py # Filesystem agent for file structure ├── core/ # Core system functionality │ ├── omniagent.py # Base agent implementation │ ├── enhanced_omniagent.py # Extended agent with customization │ └── revolutionary_omniagent.py # Unified agent with all capabilities ├── generation/ # Content generation capabilities │ ├── code_generator.py # Revolutionary code generation │ ├── doc_generator.py # Documentation generation │ └── structure_generator.py # Project structure generation ├── ui/ # User interfaces │ ├── web_ui.py # Web interface using Gradio │ ├── agent_customization_ui.py # UI for agent customization │ └── cli.py # Command-line interface ├── utils/ # Utility modules │ ├── first_principles.py # First-principles thinking implementation │ └── template_manager.py # Template management utilities └── run.py # Main entry point


## Quick Start

1. Clone the repository
2. Create a virtual environment: `python -m venv .venv`
3. Activate it: `.venv\Scripts\activate` (Windows) or `source .venv/bin/activate` (Unix)
4. Install dependencies: `pip install -r requirements.txt`
5. Launch the web UI: `python run_web_ui.py`

For detailed instructions, see [INSTALLATION.md](INSTALLATION.md).

## Usage

### Command Line

```bash
# Launch the web interface (Universal Launcher - Most Reliable)
python run_web_ui.py

# Alternative launch method via main launcher
python omnitrace/run.py --web

# Create a project from the command line
python omnitrace/run.py --project "Revolutionary App" --description "A revolutionary app that..." --revolution-level maximum

# Use a configuration file
python omnitrace/run.py --config config/config_files/default_config.yaml

For all available execution methods, see EXECUTION_METHODS.md.

Configuration Options

  • --model: LLM model to use (default: deepseek-r1:1.5b)
  • --revolution-level: Level of revolutionary thinking (moderate/high/maximum)
  • --constraint-elimination: Level of constraint elimination (cautious/moderate/aggressive)
  • --code-gen: Enable code generation
  • --no-code-gen: Disable code generation
  • --verbose: Enable verbose logging
  • --config: Path to configuration file (YAML/JSON)

Requirements

  • Python 3.9+
  • Ollama
  • Gradio
  • LangChain
  • PyYAML (for configuration files)

Agents

CEO Agent

Creates revolutionary vision by breaking down problems to fundamentals, questioning all assumptions and industry standards, and focusing on 10x improvements.

CTO Agent

Develops technical strategy that challenges conventional approaches, selects technologies based on fundamental capabilities, and designs infrastructure for exponential scaling.

Architect Agent

Designs revolutionary system architecture from first principles, challenges traditional architecture patterns, and creates novel interaction patterns between system components.

Developer Agent

Plans implementation with 10x improvement approach, questions conventional development practices, and identifies opportunities for extreme simplification.

Filesystem Agent

Creates revolutionary file structures based on fundamental relationships, challenges conventional directory organization, and designs for exponential growth.

Revolutionary First-Principles Thinking

OmnitrAIce applies Elon Musk's first-principles methodology throughout the project generation process:

  1. Breaking Down to Fundamentals: Each problem is deconstructed to its basic components
  2. Distinguishing Constraints: Separating physical limitations from artificial constraints
  3. 10x Improvement Focus: Targeting order-of-magnitude improvements rather than incremental gains
  4. Revolutionary Design: Reimagining solutions from scratch without conventional limitations
  5. Constraint Elimination: Aggressively identifying and eliminating artificial constraints

Documentation

OmnitrAIce includes comprehensive documentation:

Contributing

Contributions are welcome! Please follow standard GitHub flow (fork, branch, PR) and ensure your changes maintain the revolutionary first-principles approach.

For development guidelines, see the Development Guide.

About

OmnitrAIce project - AI-driven application for task tracing and execution

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors