StructureClaw: Traceable LLM Agents and an Executable Benchmark for Structural Engineering Workflows

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Sizhong Qin, Yi Gu, Yao Jiang, Ao Cai, Changjian Zhou, Shaoxuan Shuai, Jiachang Wang, Tianhao Shen, Yueqiang Li, Xinhao Li, Li Zeng, Yueshi Chen, Dachen Gao, Genrong Xu, Wenjie Liao, Xinzheng Lu
• Venue: arXiv: cs.SE
• Published: 2026-07-16
• Collected: 2026-07-17
• URL: http://arxiv.org/abs/2607.14896v1
• PDF: https://arxiv.org/pdf/2607.14896v1
• DOI: UNKNOWN

Plain-Language Brief

Researchers present StructureClaw, a workbench that allows Large Language Model (LLM) agents to operate through controlled engineering tasks with governed skills, typed tools, and shared artifact state. The team also introduces StructureClaw-Bench, an executable benchmark of 150 scenarios designed to test the complete evidence chain in structural engineering workflows. The study shows that the average success rate for completing these tasks increases significantly with the use of StructureClaw, but identifies two main challenges: handling invalid numerical inputs and reconstructing structural models consistently.

What the Document Says

• The paper presents StructureClaw, a workbench for LLM agents to perform structured engineering tasks with controlled skills, tools, and shared artifact state.
• StructureClaw-Bench is an executable benchmark consisting of 150 scenarios that test the complete evidence chain in structural engineering workflows.
• The benchmark evaluates the success rate of completing these tasks across ten agent-model configurations, with an average increase from 56.8% to 88.6% using StructureClaw compared to a generic-skill baseline.
• The study identifies two main challenges: handling invalid numerical inputs and reconstructing structural models consistently.

Key Concepts Mentioned

• Large Language Model (LLM)
• Structural Engineering
• Workbench
• Benchmark
• Scenarios
• Artifact-centered Evaluation

Claims Made by the Document

• Using StructureClaw significantly increases the success rate of completing structural engineering tasks.
• The study identifies two main challenges in handling numerical inputs and reconstructing structural models.

Evidence Used

• Average Success Rate rising from 56.8% to 88.6% with the full automatic workflow

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Addressing a structural-engineering request requires more than a single answer; it requires a chain of interdependent artifacts: interpreted requirements, a computable model, validation records, solver outputs, code-check records, and a final report. Evaluations centered on question answering or script generation rarely verify this complete evidence chain and may therefore reward fluent outputs even when the underlying engineering workflow is incomplete, internally inconsistent, or non-executable. To address this limitation, we present StructureClaw, an artifact-centered workbench in which LLM agents operate through governed engineering skills, typed tools, shared artifact state, and local analysis backends. We also introduce StructureClaw-Bench, an executable benchmark of 150 controlled scenarios spanning standard workflow execution, interactive robustness, and multimodal structural-model reconstruction. A scenario succeeds only when all required artifact- and execution-level assertions pass in a single run. Across ten agent-model configurations, each evaluated on the same 50 standard cases, the average Success Rate rises from 56.8% with the generic-skill baseline to 88.6% with the full automatic workflow. The interactive and multimodal evaluations identify two prominent remaining challenges: safe handling of invalid numerical inputs and fixture-consistent reconstruction of structural models. These findings show that artifact-centered evaluation can expose workflow-level failures that are difficult to identify from final responses alone, providing a more rigorous basis for evaluating and improving structural-engineering agents. The code and benchmark are available at https://github.com/structureclaw/structureclaw.