OdabNote is an 'incorrect-answer note' system that helps AI coding agents learn from their mistakes and never repeat them.
Why It Matters
Inspired by the 'odab-note' (incorrect-answer note) used by South Korean students to improve their academic performance, OdabNote brings a similar solution to AI. Instead of allowing an AI to repeatedly make the same coding mistakes, this system acts as a personal 'immune system' for each AI agent. It automatically logs specific error patterns, stores verified solutions, and then automatically checks against potential future errors, ensuring that the AI continuously learns and improves.
Who It's For
If you are a developer, software engineer, or anyone who frequently works with coding AIs like Claude, Gemini, Codex, or local models, OdabNote is the tool for you. This project aims to address the frustration of watching AI make the same mistakes over and over across different sessions. OdabNote helps you build more reliable and efficient AI agents by equipping them with the ability to learn from experience, eliminating time-consuming debugging loops.
Quick Comparison
* Manual Prompt Engineering: This approach requires users to constantly tweak and optimize prompts, lacking an automated mechanism to remember and apply lessons from past errors. * Knowledge Bases/RAG (Retrieval-Augmented Generation): These systems provide additional context or information to the AI, but they are not specialized in learning and preventing specific coding errors that the agent has previously made. * Model Fine-Tuning: This is the process of retraining the entire model on new data, which is resource- and time-intensive and lacks flexibility when handling small, specific errors or rules in an AI agent's daily workflow.
How to Get Started
To dive deeper into the architecture, implementation, and start using OdabNote to equip your coding AI with a robust 'immune system', you can directly access the project's GitHub repository.
Repo: hacker3699-max/odabnote • ?★