Agentic Specification Generator for Move Programs
This addresses the problem of under-explored verification for emerging languages like Move in smart contract development, offering a practical tool with incremental advancements over existing methods for mainstream languages.
The paper tackles the lack of specification generation tools for the Move smart contract language by introducing MSG, an automated tool that successfully generates verifiable specifications for 84% of tested functions and identifies clauses missed by experts, with improvements like a 57% increase in verifiable clauses and a 30% boost in effectiveness through agentic design and verification feedback.
While LLM-based specification generation is gaining traction, existing tools primarily focus on mainstream programming languages like C, Java, and even Solidity, leaving emerging and yet verification-oriented languages like Move underexplored. In this paper, we introduce MSG, an automated specification generation tool designed for Move smart contracts. MSG aims to highlight key insights that uniquely present when applying LLM-based specification generation to a new ecosystem. Specifically, MSG demonstrates that LLMs exhibit robust code comprehension and generation capabilities even for non-mainstream languages. MSG successfully generates verifiable specifications for 84% of tested Move functions and even identifies clauses previously overlooked by experts. Additionally, MSG shows that explicitly leveraging specification language features through an agentic, modular design improves specification quality substantially (generating 57% more verifiable clauses than conventional designs). Incorporating feedback from the verification toolchain further enhances the effectiveness of MSG, leading to a 30% increase in generated verifiable specifications.