Paper
EVMbench: Evaluating AI Agents on Smart Contract Security
Authors
Justin Wang, Andreas Bigger, Xiaohai Xu, Justin W. Lin, Andy Applebaum, Tejal Patwardhan, Alpin Yukseloglu, Olivia Watkins
Abstract
Smart contracts on public blockchains now manage large amounts of value, and vulnerabilities in these systems can lead to substantial losses. As AI agents become more capable at reading, writing, and running code, it is natural to ask how well they can already navigate this landscape, both in ways that improve security and in ways that might increase risk. We introduce EVMbench, an evaluation that measures the ability of agents to detect, patch, and exploit smart contract vulnerabilities. EVMbench draws on 117 curated vulnerabilities from 40 repositories and, in the most realistic setting, uses programmatic grading based on tests and blockchain state under a local Ethereum execution environment. We evaluate a range of frontier agents and find that they are capable of discovering and exploiting vulnerabilities end-to-end against live blockchain instances. We release code, tasks, and tooling to support continued measurement of these capabilities and future work on security.
Metadata
Related papers
Gen-Searcher: Reinforcing Agentic Search for Image Generation
Kaituo Feng, Manyuan Zhang, Shuang Chen, Yunlong Lin, Kaixuan Fan, Yilei Jian... • 2026-03-30
On-the-fly Repulsion in the Contextual Space for Rich Diversity in Diffusion Transformers
Omer Dahary, Benaya Koren, Daniel Garibi, Daniel Cohen-Or • 2026-03-30
Graphilosophy: Graph-Based Digital Humanities Computing with The Four Books
Minh-Thu Do, Quynh-Chau Le-Tran, Duc-Duy Nguyen-Mai, Thien-Trang Nguyen, Khan... • 2026-03-30
ParaSpeechCLAP: A Dual-Encoder Speech-Text Model for Rich Stylistic Language-Audio Pretraining
Anuj Diwan, Eunsol Choi, David Harwath • 2026-03-30
RAD-AI: Rethinking Architecture Documentation for AI-Augmented Ecosystems
Oliver Aleksander Larsen, Mahyar T. Moghaddam • 2026-03-30
Raw Data (Debug)
{
"raw_xml": "<entry>\n <id>http://arxiv.org/abs/2603.04915v1</id>\n <title>EVMbench: Evaluating AI Agents on Smart Contract Security</title>\n <updated>2026-03-05T07:59:14Z</updated>\n <link href='https://arxiv.org/abs/2603.04915v1' rel='alternate' type='text/html'/>\n <link href='https://arxiv.org/pdf/2603.04915v1' rel='related' title='pdf' type='application/pdf'/>\n <summary>Smart contracts on public blockchains now manage large amounts of value, and vulnerabilities in these systems can lead to substantial losses. As AI agents become more capable at reading, writing, and running code, it is natural to ask how well they can already navigate this landscape, both in ways that improve security and in ways that might increase risk. We introduce EVMbench, an evaluation that measures the ability of agents to detect, patch, and exploit smart contract vulnerabilities. EVMbench draws on 117 curated vulnerabilities from 40 repositories and, in the most realistic setting, uses programmatic grading based on tests and blockchain state under a local Ethereum execution environment. We evaluate a range of frontier agents and find that they are capable of discovering and exploiting vulnerabilities end-to-end against live blockchain instances. We release code, tasks, and tooling to support continued measurement of these capabilities and future work on security.</summary>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.CR'/>\n <published>2026-03-05T07:59:14Z</published>\n <arxiv:primary_category term='cs.LG'/>\n <author>\n <name>Justin Wang</name>\n </author>\n <author>\n <name>Andreas Bigger</name>\n </author>\n <author>\n <name>Xiaohai Xu</name>\n </author>\n <author>\n <name>Justin W. Lin</name>\n </author>\n <author>\n <name>Andy Applebaum</name>\n </author>\n <author>\n <name>Tejal Patwardhan</name>\n </author>\n <author>\n <name>Alpin Yukseloglu</name>\n </author>\n <author>\n <name>Olivia Watkins</name>\n </author>\n </entry>"
}