Research on smart contract evaluation method for open source blockchain

The smart contract can complete the agreement by automatic execution between the parties of the blockchain network. It provides the basis for various distributed application services. In recent years, the security loopholes of smart contracts have exploded frequently, caused huge economic losses. The research on its security inspection has become a hot topic and a key scientific problem that needs to be solved urgently in the field of blockchain security. In the paper, firstly, gives the vulnerability of open blockchain system smart contract, investigates the mainstream smart contract vulnerability detection methods; then proposes the vulnerability detection model of smart contract based on gated graph neural network, and tests its performance. Finally, draws a conclusion that graph neural network technology can improve the accuracy of smart contracts mainstream vulnerability testing, and plans the future research direction.