DeFiScope introduces a novel LLM-based approach to detect price manipulation attacks across diverse DeFi protocols, addressing critical gaps in current security solutions.

• Overcomes limitations of existing tools by effectively handling non-standard price models that account for 44.2% of reported attacks
• Leverages LLM reasoning capabilities to identify various manipulation patterns without requiring protocol-specific rules
• Demonstrates significant improvement over state-of-the-art systems like DeFiRanger and DeFort

This research enhances blockchain security by protecting billions in cryptocurrency assets from increasingly sophisticated attacks, creating more resilient financial infrastructure for the emerging Web3 ecosystem.

DeFiScope: Detecting Various DeFi Price Manipulations with LLM Reasoning