Skip to content

47thtechcorner/RayCodes_Ornith_1.0

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

2 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

Ornith 1.0 is HERE: The Open-Source AI Coding Agent That Fixes Its Own Code!

⚑ Ultra-Lightweight local Agentic Code Generator & Self-Repair Loop.


πŸ› οΈ Tech Stack

  • 🐍 Python 3.10+ (using urllib, json, py_compile)
  • 🎨 Streamlit (UI Dashboard)
  • πŸ€– Ollama (Ornith-1.0-9B πŸ¦‰ -> qwen2.5-coder πŸ’»)

πŸ“‚ File Structure

  • πŸ“„ app.py - Single-file app with agent logic, fallback detection, and Streamlit dashboard in under 50 lines.
  • πŸ“„ Modelfile - Custom Ollama model configuration.
  • πŸ“„ demo_output.md - Expected output logs matching model runs.
  • πŸ“„ prime_checker.py - Syntactically verified python code generated by the agent.

πŸš€ Commands (Windows PowerShell)

1. Setup Environment

python -m venv venv
.\venv\Scripts\Activate.ps1
pip install streamlit

2. Register GGUF Model in Ollama (Once downloaded)

⚠️ Note: The GGUF model files (e.g., ornith-1.0-9b-Q4_K_M.gguf) are not included in this repository. You must download the model file or clone the model source repository yourself and place the GGUF file in this directory.

ollama create ornith-1.0-9b -f Modelfile

3. Run UI Dashboard

streamlit run app.py

🌟 Features

  • πŸ” Self-Repair Loop: Automatically compile-checks code and corrects syntax errors.
  • πŸ”„ Auto-Fallback: Auto-detects local models if Ornith is still downloading.
  • πŸ“¦ Zero-Bloat: Minimal dependencies, no databases, running under 10 seconds.

Keywords

Ornith Coding Agent β€’ Ollama β€’ Streamlit β€’ Python Code Generator β€’ Self-Repair Loop β€’ Self-Healing Code β€’ Local LLM β€’ AI Agent β€’ Qwen2.5-Coder β€’ Offline AI Code Assistant

About

🦀 Ornith 1.0: An ultra-lightweight, local agentic code generator and self-repair loop built with Streamlit, Ollama, and Python.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages