Combining Mechanical and Agentic Specification Inference for Move
📰 ArXiv cs.AI
Learn to combine mechanical and agentic specification inference for Move Prover to reduce boilerplate code and improve verification efficiency
Action Steps
- Apply weakest-precondition analysis over Move bytecode to identify potential specification gaps
- Use an agentic coding CLI like Claude Code to generate specifications based on code patterns
- Combine mechanical and agentic inference results to create comprehensive specifications
- Test and refine the inferred specifications using the Move Prover
- Integrate the specification inference tool into the development workflow to reduce boilerplate code
Who Needs to Know This
This technique benefits developers and verifiers working with the Move Prover, particularly those interested in formal verification and agentic coding, as it streamlines the specification writing process and enhances code reliability.
Key Insight
💡 Combining mechanical and agentic specification inference can significantly reduce the effort required to write specifications for the Move Prover
Share This
🚀 Combine mechanical & agentic spec inference for Move Prover to boost verification efficiency! 📈
Full Article
Title: Combining Mechanical and Agentic Specification Inference for Move
Abstract:
arXiv:2605.10005v1 Announce Type: cross Abstract: In this paper, we describe early work on a specification inference tool for the Move Prover that combines a weakest-precondition (WP) analysis over Move bytecode with an agentic coding CLI such as Claude Code. Specification inference reduces the boilerplate of writing specifications in Move: in order to verify a high-level property such as a global state invariant, pre- and post-conditions for the supporting functions typically have to be written
Abstract:
arXiv:2605.10005v1 Announce Type: cross Abstract: In this paper, we describe early work on a specification inference tool for the Move Prover that combines a weakest-precondition (WP) analysis over Move bytecode with an agentic coding CLI such as Claude Code. Specification inference reduces the boilerplate of writing specifications in Move: in order to verify a high-level property such as a global state invariant, pre- and post-conditions for the supporting functions typically have to be written
DeepCamp AI