VASPilot: MCP-facilitated multi-agent intelligence for autonomous VASP simulations
Abstract
Abstract Density-functional-theory (DFT) simulations with the Vienna Ab initio Simulation Package (VASP) are indispensable in computational materials science but often require extensive manual setup, monitoring, and postprocessing. Here, we introduce VASPilot, an open-source platform that fully automates VASP workflows via a multi-agent architecture built on the CrewAI framework and a standardized model context protocol (MCP). VASPilot’s agent suite handles every stage of a VASP study from retrieving crystal structures and generating input files to submitting Slurm jobs, parsing error messages, and dynamically adjusting parameters for seamless restarts. A lightweight Quart-based web interface provides intuitive task submission, real-time progress tracking, and drill-down access to execution logs, structure visualizations, and plots. We validated VASPilot on both routine and advanced benchmarks: automated band-structure and density-of-states calculations (including on-the-fly symmetry corrections), plane-wave cutoff convergence tests, lattice-constant optimizations with various van der Waals corrections, and cross-material band-gap comparisons for transition-metal dichalcogenides. In all cases, VASPilot completed the missions reliably and without manual intervention. Moreover, its modular design allows easy extension to other DFT codes simply by deploying the appropriate MCP server. By offloading technical overhead, VASPilot enables researchers to focus on scientific discovery and accelerates high-throughput computational materials research.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
VASPilot autonomously configured and executed VASP studies across convergence, band-structure, and lattice-optimization tasks.
“VASPilot autonomously configured and executed VASP studies across convergence, band-structure, and lattice-optimization tasks [25].”
From From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery (2026) · cites this paper as [25]
The remaining literature was distributed across general materials-workflow papers, atomistic simulation infrastructures, MOF and porous-material screening, and polymer simulation.
“The remaining literature was distributed across general materials-workflow papers, atomistic simulation infrastructures, MOF and porous-material screening, and polymer simulation [17,21,22,24-26,32,35-42].”
From From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery (2026) · cites this paper as [25]
These papers typically split the workflow into role-specific agents for planning, structure retrieval, simulation execution, analysis, and quality control.
“These papers typically split the workflow into role-specific agents for planning, structure retrieval, simulation execution, analysis, and quality control [18,19,21,25,26,35-38,41,42].”
From From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery (2026) · cites this paper as [25]
Some studies are simulator-native: the agent configures or executes atomistic or thermodynamic tools directly.
“Some studies are simulator-native: the agent configures or executes atomistic or thermodynamic tools directly [21,25,28,29,31,33,36,38-42].”
From From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery (2026) · cites this paper as [25]
In several papers, the strongest reproducibility evidence came from public code, tutorials, datasets, or supplementary traces.
“In several papers, the strongest reproducibility evidence came from public code, tutorials, datasets, or supplementary traces [18-21,23,25,28,30,31,35-42].”
From From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery (2026) · cites this paper as [25]
The most persuasive papers linked the language model to DFT, molecular dynamics, adsorption screening, process simulation, CALPHAD, crystallographic curation, or robotic execution.
“Instead, they linked it to DFT, molecular dynamics, adsorption screening, process simulation, CALPHAD, crystallographic curation, or robotic execution [18,21,25,28,29,31-36,39-42].”
From From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery (2026) · cites this paper as [25]
Papers such as SmartCIF, DynaMate, VASPilot, Matty, GENIUS, and ToPolyAgent show that a substantial share of the current advance lies not in replacing scientific judgment with text generation, but in making brittle infrastructural steps programmable and recoverable.
“Papers such as SmartCIF, DynaMate, VASPilot, Matty, GENIUS, and ToPolyAgent show that a substantial share of the current advance lies not in replacing scientific judgment with text generation, but in making these brittle infrastructural steps programmable and recoverable [21,25,37-39,42].”
From From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery (2026) · cites this paper as [25]
The corpus already demonstrates several concrete successes: autonomous VASP execution across routine electronic-structure tasks, large-scale Quantum ESPRESSO protocol generation and repair, substantial reduction of adsorption-search effort without collapsing energy quality, adaptive alloy optimization under repeated MD feedback, catalyst discovery with experimental confirmation, and one genuine多智能
“Importantly, the corpus already demonstrates several concrete successes: autonomous VASP execution across routine electronic-structure tasks [25], large-scale Quantum ESPRESSO protocol generation and repair [42], substantial reduction of adsorption-search effort without collapsing energy quality [28], adaptive alloy optimization under repeated MD feedback [33], catalyst discovery with experimental”
From From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery (2026) · cites this paper as [25]
The citation lists haven’t been fetched yet
References 50 · Cited by 7
- Authors
- Liu 刘, J. 家., Zhu 朱, T. 天., Ye 叶, C. 财., Fang 方, Z. 忠., Weng 翁, H. 红., & Wu 吴, Q. 泉.
- Venue
- Chinese Physics B
- Publisher
- IOP Publishing
- Published
- 2025-11-01
- Volume
- 34
- Issue
- 11
- Pages
- 117106
- ISSN
- 1674-1056 (ISSN-L)
- 2058-3834
- OpenAlex
- W7081975031
- Cited by
- 7
- References
- 50
- Citations by year
- 2026: 7
- FWCI
- 1.47
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- Chinese Physics B (publishedVersion)
- Topics
- Physical Sciences › Materials Science › Materials Chemistry › Machine Learning in Materials Science
- Physical Sciences › Physics and Astronomy › Atomic and Molecular Physics, and Optics › Advanced Chemical Physics Studies
- Social Sciences › Decision Sciences › Information Systems and Management › Scientific Computing and Data Management
- Keywords
- multi-agent architecture, CrewAI, Model Context Protocol, density functional theory, SLURM job submission, van der Waals corrections, transition metal dichalcogenides, electronic band structure, density of states, open source software
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
