Skip to content

Detecting Reentrancy Vulnerability in Smart Contracts using Graph Convolution Networks

License

Notifications You must be signed in to change notification settings

Hozefa-L/HARDEN

Repository files navigation

HARDEN

Detecting Reentrancy Vulnerability in Smart Contracts using Graph Convolution Networks

This project's aim was to detect reentrancy vulnerability in Ethereum Smart Contracts. Contains 342 smart contracts .dot dataset used for training and testing HARDEN Algorithm.

Below are important libraries required to implement HARDEN Algorithm in python: DGL (Deep Graph Library) - Used to Load, train and test Graph Neural Network Model Pytorch - backend environment for computation Networkx - Can be used to manipulate graphs Ethersolve - Takes bytecode as input and returns sound Control Flow graphs

#Reentrancy_dataset contains smart contract dot graphs dataset used for reentrancy detection

#HARDEN_Bytecode_Graph_Generation can be used to convert smart contracts to evm bytecodes to dot Control Flow graphs

#HARDEN_Load_Test_Train_GCN loads graph dataset and GCN model using DGL to train and test GCN for reentrancy detection in Ethereum Smart Contracts.

About

Detecting Reentrancy Vulnerability in Smart Contracts using Graph Convolution Networks

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages