{
  "abstract": "Computer-aided engineering (CAE) simulation is among the largest and most demanding areas of engineering, where setting up a solver such as OpenFOAM, FEniCS, or COMSOL takes real expertise. Large language model (LLM) agents promise to turn a natural-language request into a working simulation, and recent CAE agents add simulation-specific machinery: multi-agent decomposition, domain retrieval, and scripted reflection. That machinery suited weak base models; modern harnesses already supply multi-turn reasoning, tool use, and execution feedback. We ask what a CAE simulation agent still needs beyond a generic harness. With information access and repair budget held fixed, a single-agent harness matches or beats multi-agent specialized systems (FoamBench 96.4\\% vs.\\ 88.2\\%). Ablations trace this to capabilities the harness already provides: execution-feedback repair lifts FoamBench from 71.8\\% with no repair round to 96.4\\%, while scripted reflection adds nothing. The one input that still helps is domain knowledge supplied as solver tutorials, our largest measured gain (80.9\\% to 96.4\\%).",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.03718",
  "author_total": 2,
  "authors": [
    {
      "name": "Jiasheng Shi"
    },
    {
      "name": "Tianhan Zhang"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/4j00wbp49d1f4.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.03718",
  "fulltext_url": null,
  "id": "4j00wbp49d1f4",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.03718",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.CE"
    ],
    [
      "arxiv",
      "cs.CL"
    ],
    [
      "arxiv",
      "physics.comp-ph"
    ],
    [
      "openalex",
      "1702|Artificial Intelligence Applications"
    ]
  ],
  "openalex": "W7208710705",
  "pdf_url": "https://arxiv.org/pdf/2609.03718",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-03",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/4j00wbp49d1f4.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:18:31Z",
      "native_id": "2609.03718",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.03718"
    }
  ],
  "subjects": [
    {
      "field": "physics",
      "origin": "arxiv:physics.comp-ph",
      "subfield": "computational"
    }
  ],
  "title": "What Do CAE Simulation Agents Really Need Beyond a Generic Harness?",
  "title_text": "What Do CAE Simulation Agents Really Need Beyond a Generic Harness?",
  "type": "preprint",
  "updated_at": "2026-10-30T12:18:31Z",
  "url": "https://research.draconitic.com/paper/4j00wbp49d1f4",
  "venue": "arXiv",
  "versions": []
}