Metraipolis › archive › dc_7d276ddc315f

Demystifying Agent Skills for Smart Contract Auditing: Design, Effectiveness, Behavioral Impact — arXiv:2609.29454

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.29454
License (as archived)
CC0-1.0
Archived
2026-10-08T06:01:52.179Z
SHA-256 of the archived text
fdfa511f526fdb944b04c565bf798fe00be8e496769066d3d9f11bb530ce454e
Stable URL of this page
https://metraipolis.com/documento/dc_7d276ddc315f

How to cite

Demystifying Agent Skills for Smart Contract Auditing: Design, Effectiveness, Behavioral Impact — arXiv:2609.29454 — arXiv (en), https://arxiv.org/abs/2609.29454 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7d276ddc315f (Metraipolis archive dc_7d276ddc315f, sha256 fdfa511f526fdb944b04c565bf798fe00be8e496769066d3d9f11bb530ce454e)
BibTeX
@misc{dc7d276ddc315f,
  title = {Demystifying Agent Skills for Smart Contract Auditing: Design, Effectiveness, Behavioral Impact — arXiv:2609.29454},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.29454},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7d276ddc315f; sha256 fdfa511f526fdb944b04c565bf798fe00be8e496769066d3d9f11bb530ce454e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.29454. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LLM agents, notably Claude Code and OpenAI Codex, are emerging as versatile tools beyond coding agents only. These agents can be enhanced with skills---reusable artifacts that package domain knowledge, workflows, and tool-use instructions. To date, however, little is known about how such skills are designed or how they affect agent effectiveness and behavior in practice. In this paper, we investigate these questions in smart contract security auditing, a domain in which agents have shown substantial promise. We systematically collect 83 smart contract audit skills from the wild and evaluate them on EVMBench across seven agent--model configurations. Our study examines three dimensions: (i) the design characteristics of audit skills, including their structure, knowledge representations, workflows, and tool dependencies; (ii) their effectiveness in improving vulnerability detection; and (iii) their influence on agent execution trajectories. We find that audit skills are mostly lightweight but heterogeneous in design, covering a broad yet imbalanced range of vulnerability types. Their effectiveness is determined primarily by the model rather than the agent harness: Codex/GPT-5.5 achieves the largest gains, improving detection score by 22.8% and captured award by 43.2%. We further find that skill triggering is a key bottleneck. When triggered, skills preserve a shared six-stage audit workflow while exhibiting distinct loading patterns and differential effects on agent behavior across configurations. We release our skill corpus and artifacts to support future research.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-25 in cs.SE).
Categorie arXiv: Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2609.29454 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T05:25:51.962Z dalla pagina https://arxiv.org/catchup/cs.SE/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).