Saved in:
Bibliographic Details
Main Authors: Goel, Abhinav, Shah, Chaitya, Capponi, Agostino, Gliozzo, Alfio
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.13808
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912904502050816
author Goel, Abhinav
Shah, Chaitya
Capponi, Agostino
Gliozzo, Alfio
author_facet Goel, Abhinav
Shah, Chaitya
Capponi, Agostino
Gliozzo, Alfio
contents We present an end-to-end framework for systematic evaluation of LLM-generated smart contracts from natural-language specifications. The system parses contractual text into structured schemas, generates Solidity code, and performs automated quality assessment through compilation and security checks. Using CrewAI-style agent teams with iterative refinement, the pipeline produces structured artifacts with full provenance metadata. Quality is measured across five dimensions, including functional completeness, variable fidelity, state-machine correctness, business-logic fidelity, and code quality aggregated into composite scores. The framework supports paired evaluation against ground-truth implementations, quantifying alignment and identifying systematic error modes such as logic omissions and state transition inconsistencies. This provides a reproducible benchmark for empirical research on smart contract synthesis quality and supports extensions to formal verification and compliance checking.
format Preprint
id arxiv_https___arxiv_org_abs_2602_13808
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle An end-to-end agentic pipeline for smart contract translation and quality evaluation
Goel, Abhinav
Shah, Chaitya
Capponi, Agostino
Gliozzo, Alfio
Artificial Intelligence
Software Engineering
D.3; I.2; J.6; D.2
We present an end-to-end framework for systematic evaluation of LLM-generated smart contracts from natural-language specifications. The system parses contractual text into structured schemas, generates Solidity code, and performs automated quality assessment through compilation and security checks. Using CrewAI-style agent teams with iterative refinement, the pipeline produces structured artifacts with full provenance metadata. Quality is measured across five dimensions, including functional completeness, variable fidelity, state-machine correctness, business-logic fidelity, and code quality aggregated into composite scores. The framework supports paired evaluation against ground-truth implementations, quantifying alignment and identifying systematic error modes such as logic omissions and state transition inconsistencies. This provides a reproducible benchmark for empirical research on smart contract synthesis quality and supports extensions to formal verification and compliance checking.
title An end-to-end agentic pipeline for smart contract translation and quality evaluation
topic Artificial Intelligence
Software Engineering
D.3; I.2; J.6; D.2
url https://arxiv.org/abs/2602.13808