topas-editor and TOPilot: tools for AI-driven analysis of diffraction and total scattering data
Abstract
This paper describes topas-editor and TOPilot, which provide a powerful interface to the TOPAS software suite and allow generative AI-driven automation of complex crystallographic tasks. topas-editor is a TOPAS-aware VS Code extension designed for the rapid creation and editing of TOPAS input files and for interacting with the software. It knows TOPAS syntax, has crystallographic-aware editing features and provides in-line user guidance via hover tips and automated links to reference manuals. TOPilot is a collection of Python routines and machine-accessible reference documents that allow AI-driven automated analysis in TOPAS. It uses a transferable topas-inp-writer skill that can be applied in all TOPAS application areas. We demonstrate the power and reliability of the approach using examples including: a complex Rietveld refinement with sample- and data-related complications; automatic analysis of multiple data sets; a pipeline from indexing through structure solution to refinement; automatic identification of the phases present in a sample; single crystal and powder diffraction analysis of incommensurate structures; and fitting PDF data. TOPilot can be used directly alongside topas-editor, via a stand-alone topilot-wizard, or via command line interfaces or desktop apps from major AI providers. TOPilot was developed and tested using Claude large language models, but can be configured to use other cloud or local models. topas-editor and TOPilot are freely available.
Related
- A search for non-Newtonian force in nm range from slow neutron scattering by noble gases and neutron diffraction on atom pairs
- Double diffraction production for dipole-dipole scattering at high energies
- Exactly Solvable Diffraction-Grating Scattering Problems for TE and TM waves
- Water, vacancies, and competing exchange interactions in Prussian blue analogues: a neutron diffraction study of field and dehydration-driven magnetic transitions
- Microfacet-based physics-driven framework for single-photon LiDAR rough-surface scattering and detection
BibTeX
@misc{evans2026topaseditor,
title = {{topas-editor and TOPilot: tools for AI-driven analysis of diffraction and total scattering data}},
author = {John S. O. Evans and Oliver J. Wagstaff and Ivana R. Evans and Arnt Kern and Alan A. Coelho},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.23512},
eprint = {2610.23512},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.23512},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - topas-editor and TOPilot: tools for AI-driven analysis of diffraction and total scattering data
AU - John S. O. Evans
AU - Oliver J. Wagstaff
AU - Ivana R. Evans
AU - Arnt Kern
AU - Alan A. Coelho
PY - 2026
DA - 2026/10/16
PB - arXiv
DO - 10.48550/arxiv.2610.23512
UR - https://arxiv.org/abs/2610.23512
AB - This paper describes topas-editor and TOPilot, which provide a powerful interface to the TOPAS software suite and allow generative AI-driven automation of complex crystallographic tasks. topas-editor is a TOPAS-aware VS Code extension designed for the rapid creation and editing of TOPAS input files and for interacting with the software. It knows TOPAS syntax, has crystallographic-aware editing features and provides in-line user guidance via hover tips and automated links to reference manuals. TOPilot is a collection of Python routines and machine-accessible reference documents that allow AI-driven automated analysis in TOPAS. It uses a transferable topas-inp-writer skill that can be applied in all TOPAS application areas. We demonstrate the power and reliability of the approach using examples including: a complex Rietveld refinement with sample- and data-related complications; automatic analysis of multiple data sets; a pipeline from indexing through structure solution to refinement; automatic identification of the phases present in a sample; single crystal and powder diffraction analysis of incommensurate structures; and fitting PDF data. TOPilot can be used directly alongside topas-editor, via a stand-alone topilot-wizard, or via command line interfaces or desktop apps from major AI providers. TOPilot was developed and tested using Claude large language models, but can be configured to use other cloud or local models. topas-editor and TOPilot are freely available.
ER -