While ensuring the safety of smart contracts during the development stage is crucial, formalizing natural language requirements for their verification can be challenging and time-consuming. When it comes to standard-based contracts, Ethereum provides natural language guidelines to ensure compliant and safe implementations. Deviating from the intended standard behavior can lead to serious security pitfalls. To address this issue, we propose a verification approach for ensuring the safety of ERC-based contracts using model checking. Our approach employs Mistral as a Large Language Model (LLM) to generate Computational Tree Logic (CTL) specifications from natural language requirements. We demonstrate the applicability of our approach through an ERC-20 smart contract.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Towards an Automated Verification Approach for ERC-Based Smart Contracts

  • Rim Ben Fekih,
  • Mariam Lahami,
  • Mohamed Salem El Eze,
  • Salma Bradai,
  • Mohamed Jmaiel

摘要

While ensuring the safety of smart contracts during the development stage is crucial, formalizing natural language requirements for their verification can be challenging and time-consuming. When it comes to standard-based contracts, Ethereum provides natural language guidelines to ensure compliant and safe implementations. Deviating from the intended standard behavior can lead to serious security pitfalls. To address this issue, we propose a verification approach for ensuring the safety of ERC-based contracts using model checking. Our approach employs Mistral as a Large Language Model (LLM) to generate Computational Tree Logic (CTL) specifications from natural language requirements. We demonstrate the applicability of our approach through an ERC-20 smart contract.