{
  "abstract": "Large language models (LLMs) can drive scientific software through function calling, but free-form tool use is fragile on long, data-dependent processing chains. Models omit arguments, invent units, feed the previous artifact into the wrong input, and report completion without a verifiable product. We present GeoAgent, an agent system for seismic data processing built on Madagascar's roughly 1{,}965 command-line programs. GeoAgent exposes 54 operator agents (51 generated from a declarative specification), retrieves from an index of 1{,}937 operators, and encodes the processing flow as a nine-stage state graph. Four explicit contracts---named artifacts, input resolution, parameter takeover, and artifact validation---move everything the graph can determine out of the model's discretion. A deterministic geometry tool parses the self-describing SPS/RPS/XPS and SEG-Y headers, cross-checks eight consistency conditions, and reproduces the reference shell recipe exactly. The system runs in containers with session- and task-level isolation, bounded execution, and defenses against silent failures. On a 2-D land dataset, the nine-stage pipeline completes in about six to eleven minutes and yields a verified artifact at every stage, while retrieval reduces per-turn tool context from about 12k to 2k tokens. We report the design, deployment experience, and motivating failure modes.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.31096",
  "author_total": 7,
  "authors": [
    {
      "name": "Zhilong Fang"
    },
    {
      "name": "Qi Zhang"
    },
    {
      "name": "Jujie Yan"
    },
    {
      "name": "Zhong Wang"
    },
    {
      "name": "Xinbin Gu"
    },
    {
      "name": "Ziyang Zhang"
    },
    {
      "name": "Hua Wang"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/ft7q7ndc7f3xn.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.31096",
  "fulltext_url": null,
  "id": "ft7q7ndc7f3xn",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.31096",
  "language": null,
  "licence": "CC-BY-NC-ND-4.0",
  "licence_label": "CC BY-NC-ND 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by-nc-nd/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "physics.geo-ph"
    ],
    [
      "openalex",
      "1802|Scientific Computing and Data Management"
    ]
  ],
  "openalex": "W7223400071",
  "pdf_url": "https://arxiv.org/pdf/2610.31096",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-23",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/ft7q7ndc7f3xn.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:35:15Z",
      "native_id": "2610.31096",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.31096"
    }
  ],
  "subjects": [
    {
      "field": "earth",
      "origin": "arxiv:physics.geo-ph",
      "subfield": "geophysics"
    }
  ],
  "title": "GeoAgent: Contract-Constrained LLM Agents for Seismic Data Processing",
  "title_text": "GeoAgent: Contract-Constrained LLM Agents for Seismic Data Processing",
  "type": "preprint",
  "updated_at": "2026-10-30T12:35:15Z",
  "url": "https://research.draconitic.com/paper/ft7q7ndc7f3xn",
  "venue": "arXiv",
  "versions": []
}