SmartIntentNN: Towards Smart Contract Intent Detection

Résumé : Researchers currently have been focusing on smart contract vulnerability detection, but we find that developers' intent to write smart contracts is a more noteworthy security concern because smart contracts with malicious intent have caused significant financial loss to users. A more unfortunate fact is that we can only rely on manual audits to check for unfriendly smart contracts. In this paper, we propose \textsc{SmartIntentNN}, Smart Contract Intent Neural Network, a deep learning-based tool that aims to automate the process of developers' intent detection in smart contracts, saving human resources and overhead. The demo video is available on \url{https://youtu.be/ho1SMtYm-wI}.