Historical analysis of decentralized applications

• Main Author: Ng, Teng Hian
• Other Authors: Li Yi (SCSE)
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science; Smart contract; Blockchain; Decentralized application; Analysis
• Online Access: https://hdl.handle.net/10356/181134
• Institution: Nanyang Technological University

This report explores the development and application of automated tools and frameworks for analyzing Ethereum-based smart contracts, particularly within the decentralized finance (DeFi) sector. Smart contracts, which enable self-executing agreements, present unique challenges in terms of security and efficiency due to their immutable nature once deployed. The primary objective of this study is to develop methods that enhance the functionality of existing program analysis tools, such as InvCon, Diligent Fuzzing, Mythril, and Foundry, to provide a comprehensive understanding of smart contract behavior and to identify vulnerabilities and inefficiencies. The study outlines the use of InvCon for testing contract invariants, which ensures that smart contracts behave as expected under various conditions. Diligent Fuzzing is leveraged to uncover potential vulnerabilities in smart contracts by automatically generating and testing random inputs. Mythril is used to perform symbolic execution and taint analysis, identifying common security issues in smart contract code. Foundry is introduced as a powerful toolchain for managing the writing, testing, and deployment of smart contracts, streamlining the entire development lifecycle. Additionally, the concept of proxy contracts is explored, highlighting their importance in enabling contract upgradeability while preserving state. While the study encountered challenges, such as compatibility issues with older Solidity versions and the complexity of blockchain technology, it successfully establishes a foundation for future advancements in smart contract analysis. The findings illustrate the potential for automated testing to improve the security and functionality of decentralized applications. This report also provides recommendations for further research, including cross-chain compatibility, gas optimization, and the integration of AI-driven analysis frameworks. The methodologies developed here pave the way for more robust security practices in the blockchain ecosystem.