Skip to content

mrnetwork0001/codecrib

Repository files navigation

CodeCrib: The ELI5 Smart Contract Reader

06 03 2026_23 11 31_REC

CodeCrib is a decentralized AI (dAI) educational platform designed to make Web3 accessible to everyone. By leveraging OpenGradient’s verifiable AI, the app automatically pulls complex smart contracts and translates them into simple, "Explain Like I'm 5" (ELI5) summaries.

View Project

Features

Daily Insights: Automatically fetch and explain a new popular smart contract every 24 hours.

Verifiable AI: Powered by OpenGradient's Secure LLM Inference to ensure explanations are untampered and secure.

Interactive UI: A minimalist, flashcard-style interface built for readability and engagement.

Safety Checks: Every explanation includes a dedicated "Is it safe?" section to help beginners understand contract risks.

Tech Stack

Intelligence: OpenGradient SDK (Secure TEE-based LLM Inference).

Frontend: Lovable (React + Tailwind CSS).

Backend: Python 3.12 (Running as a scheduled cron worker on a VPS).

Database: Supabase (Real-time data synchronization).

Getting Started

Prerequisites Python 3.10+ installed on your VPS.

An OpenGradient Account and $OPG tokens (Base Sepolia).

A Supabase project URL and API key.

Installation

  1. Clone the repository:
git clone https://github.com/yourusername/codecrib.git
cd codecrib
  1. Set up the Virtual Environment:
python3 -m venv venv
source venv/bin/activate
pip install opengradient supabase python-dotenv
  1. Configure Environment Variables: Create a .env file in the root directory:
OG_PRIVATE_KEY=your_private_key
SUPABASE_URL=your_supabase_url
SUPABASE_KEY=your_supabase_anon_key

How It Works

The Worker: A Python script runs daily via Cron on the VPS.

Inference: It sends contract code to OpenGradient's TEE-protected GPT-4o model.

Storage: The resulting "ELI5" summary is stored in a Supabase table.

Display: The Lovable frontend listens for database updates and renders the latest lesson instantly.

Security & Privacy

CodeCrib uses Trusted Execution Environments (TEEs) via OpenGradient. This means the AI processing your code is mathematically verifiable and protected from external tampering.

License

This project is licensed under the MIT License - see the LICENSE.md file for details.

About

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages