On the Efficacy of Smart Contract Analysis Tools

Abstract

Distributed Ledger Technologies are an emerging reality opening the way to new application design paradigms like smart contracts-based distributed applications. If on one side they are creating new markets and opportunities, on the other they are exposing users to new security issues deriving from the scarce maturity in terms of security practices in their design and development. This paper raises a warning about the efficacy of a state-of-the-art software testing tool, namely Mythril, by challenging it with real smart contracts extracted from the Code4arena competitions and comparing its performance with security audits released during the contests. The paper highlights possible root causes of inefficiency, opening the way toward more scalable and efficient smart contract testing tools.