GeoAgent: Contract-Constrained LLM Agents for Seismic Data Processing
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.
Related
- SPBench: A Multi-Task Evaluation Benchmark for Exploration Seismic Processing
- GeoAgent: An Agent-Based QGIS Plugin for Natural-Language-Driven Geospatial Analysis with Multi-Backend Large Language Model Support
- Minor Risk to Contract Metazoan Parasites via the Consumption of Edible Snails, Born and Reared in France
- Impact of proposed wind turbines on seismic monitoring stations at the Xiaolangdi Reservoir
- Seismic imaging of the crust and upper mantle beneath Taiwan using full-waveform inversion of teleseismic body waves
BibTeX
@misc{fang2026geoagent,
title = {{GeoAgent: Contract-Constrained LLM Agents for Seismic Data Processing}},
author = {Zhilong Fang and Qi Zhang and Jujie Yan and Zhong Wang and Xinbin Gu and Ziyang Zhang and Hua Wang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.31096},
eprint = {2610.31096},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.31096},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - GeoAgent: Contract-Constrained LLM Agents for Seismic Data Processing
AU - Zhilong Fang
AU - Qi Zhang
AU - Jujie Yan
AU - Zhong Wang
AU - Xinbin Gu
AU - Ziyang Zhang
AU - Hua Wang
PY - 2026
DA - 2026/10/23
PB - arXiv
DO - 10.48550/arxiv.2610.31096
UR - https://arxiv.org/abs/2610.31096
AB - 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.
ER -