VASPilot: MCP-facilitated multi-agent intelligence for autonomous VASP simulations

doi:10.1088/1674-1056/ae0681

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.

Keywords: VASP agent model context protocol (MCP) VASPilot

Received: 25 August 2025
Revised: 09 September 2025
Accepted manuscript online: 15 September 2025

PACS:

71.15.Mb

(Density functional theory, local density approximation, gradient and other corrections)

Fund: Project supported by the Science Center of the National Natural Science Foundation of China (Grant No. 12188101), the National Key R&D Program of China (Grant Nos. 2023YFA1607400 and 2022YFA1403800), the National Natural Science Foundation of China (Grant Nos. 12274436, 11925408, and 11921004), and the New Cornerstone Science Foundation through the XPLORER PRIZE. The AI-driven experiments, simulations and model training were performed on the robotic AI-Scientist platform of the Chinese Academy of Science.

Corresponding Authors: Hongming Weng, Quansheng Wu
E-mail: hmweng@iphy.ac.cn;quansheng.wu@iphy.ac.cn

Cite this article:

Jiaxuan Liu(刘家轩), Tiannian Zhu(朱天念), Caiyuan Ye(叶财渊), Zhong Fang(方忠), Hongming Weng(翁红明), and Quansheng Wu(吴泉生) VASPilot: MCP-facilitated multi-agent intelligence for autonomous VASP simulations 2025 Chin. Phys. B 34 117106

[1] Xiao R, Li H and Chen L 2012 Chem. Mater. 24 4242
[2] Wang Y, Yu X, Xu S, Bai J, Xiao R, Hu Y S, Li H, Yang X Q, Chen L and Huang X 2013 Nat. Commun. 4 2365
[3] Wang X, Xiao R, Li H and Chen L 2017 Phys. Rev. Lett. 118 195901
[4] Jia Y, Yu J, Liu J, Herzog-Arbeitman J, Qi Z, Pi H, Regnault N, Weng H, Bernevig B A and Wu Q 2024 Phys. Rev. B 109 205121
[5] Wang C, Zhang X W, Liu X, He Y, Xu X, Ran Y, Cao T and Xiao D 2024 Phys. Rev. Lett. 132 036501
[6] Devakul T, Crpel V, Zhang Y and Fu L 2021 Nat. Commun. 12 6730
[7] Zhang S, Wu Q, Liu Y and Yazyev O V 2019 Phys. Rev. B 99 035142
[8] Liu Z, Zhang S, Fang Z, Weng H and Wu Q 2021 Phys. Rev. Res. 6 043185
[9] Pi H, Zhang S, Xu Y, Fang Z, Weng H and Wu Q 2024 npj Comput. Mater. 10 276
[10] Kresse G and Furthmller J 1996 Phys. Rev. B 54 11169
[11] Giannozzi P, Baroni S, Bonini N, et al. 2009 J. Phys.: Condens. Matter 21 395502
[12] Giannozzi P, Andreussi O, Brumme T, et al. 2017 J. Phys.: Condens. Matter 29 465901
[13] Clark S J, Segall M D, Pickard C J, Hasnip P J, Probert M I J, Refson K and Payne M C 2005 Z. Kristallogr. 220 567
[14] Ozaki T 2003 Phys. Rev. B 67 155108
[15] Weng H, Ozaki T and Terakura K 2009 Phys. Rev. B 79 235118
[16] Soler J M, Artacho E, Gale J D, Garca A, Junquera J, Ordejn P and Snchez-Portal D 2002 J. Phys.: Condens. Matter 14 2745
[17] Blaha P, Schwarz K, Tran F, Laskowski R, Madsen G K H and Marks L D 2020 J. Chem. Phys. 152 074101
[18] The Elk Code URL http://elk.sourceforge.net/
[19] Wang V, Xu N, Liu J C, Tang G and Geng W T 2021 Comput. Phys. Commun. 267 108033
[20] Hjorth Larsen A, Jrgen Mortensen J, Blomqvist J, et al. 2017 J. Phys.: Condens. Matter 29 273002
[21] Ong S P, RichardsWD, Jain A, Hautier G, Kocher M, Cholia S, Gunter D, Chevrier V L, Persson K A and Ceder G 2013 Comput. Mater. Sci. 68 314
[22] Wang H, Zhang L, Han J and E W 2018 Comput. Phys. Commun. 228 178
[23] Batzner S, Musaelian A, Sun L, et al. 2022 Nat. Commun. 13 2453
[24] Deng B, Zhong P, Jun K, Riebesell J, Han K, Bartel C J and Ceder G 2023 Nat. Mach. Intell. 5 1031
[25] Yang A, Li A, Yang B, et al. 2025 arXiv:2505.09388 [cs.CL]
[26] DeepSeek-AI, Liu A, Feng B, et al. 2024 arXiv:2412.19437 [cs.CL]
[27] DeepSeek-AI, Guo D, Yang D, et al. 2025 arXiv:2501.12948 [cs.CL]
[28] Team G V, Hong W, Yu W, et al. 2025 arXiv:2507.01006 [cs.CV]
[29] Meta AI 2025 Llama 4: A new era in multimodal intelligence
[30] Team G, Kamath A, Ferret J, et al. 2025 arXiv:2503.19786 [cs.CL]
[31] Kim S, Jung Y and Schrier J 2024 J. Am. Chem. Soc. 146 19654
[32] Song Z, Lu S, Ju M, Zhou Q andWang J 2024 arXiv:2407.07016 [condmat. mtrl-sci]
[33] Zheng Z, Zhang O, Borgs C, Chayes J T and Yaghi O M 2023 J. Am. Chem. Soc. 145 18048
[34] Lee S, Cruse K, Baibakova V, Ceder G and Jain A 2025 Text-mined dataset of solid-state syntheses
[35] Wang X D, Chen Z R, Guo P J, Gao Z F, Mu C and Lu Z Y 2025 arXiv:2507.16307 [cs.LG]
[36] Hong S, Zhuge M, Chen J, et al. 2024 MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework
[37] LangChain AI 2023 LangGraph: Build resilient language agents as graphs
[38] CrewAI 2024 CrewAI: The Leading Multi-Agent Platform
[39] Ghafarollahi A and Buehler M J 2024 arXiv:2407.10022 [cs.AI]
[40] Thompson A P, Aktulga H M, Berger R, et al. 2022 Comput. Phys. Commun. 271 108171
[41] Zhang B, Li X, Xu H, Jin Z, Wu Q and Li C 2025 arXiv:2507.04053 [cond-mat.mtrl-sci]
[42] Han X Q, Gao Z F, Guo P J and Lu Z Y 2025 PhysAgent: A Multi-Agent Approach to Automated Discovery of Physical Laws
[43] Ni Z, Li Y, Hu K, et al. 2024 arXiv:2411.08063 [physics.soc-ph]
[44] Song T, Luo M, Zhang X, et al. 2025 J. Am. Chem. Soc. 147 12534
[45] Zou Y, Cheng A H, Aldossary A, et al. 2025 Matter 8 102263
[46] Wang Z, Huang H, Zhao H, Xu C, Zhu S, Janssen J and Viswanathan V 2025 arXiv:2507.14267 [cs.AI]
[47] Bannwarth C, Caldeweyher E, Ehlert S, et al. 2021 WIREs Comput. Mol. Sci. 11 e1493
[48] Model Context Protocol 2024 Model context protocol
[49] Jain A, Ong S P, Hautier G, et al. 2013 APL Mater. 1 011002
[50] Lewis P, Perez E, Piktus A, et al. 2020 arXiv:2005.11401 [cs.CL]
[51] Pallets Projects 2025 Quart
[52] Pallets Projects 2025 Flask
[53] Grimme S 2006 J. Comput. Chem. 27 1787
[54] Grimme S, Antony J, Ehrlich S and Krieg H 2010 J. Chem. Phys. 132 154104
[55] Grimme S, Ehrlich S and Goerigk L 2011 J. Comput. Chem. 32 1456
[56] Tkatchenko A and Scheffler M 2009 Phys. Rev. Lett. 102 073005
[57] Klime J, Bowler D R and Michaelides A 2011 Phys. Rev. B 83 195131
[58] Klime J, Bowler D R and Michaelides A 2010 J. Phys.: Condens. Matter 22 022201
[59] Peng H, Yang Z H, Perdew J P and Sun J 2016 Phys. Rev. X 6 041005
[60] Ning J, Kothakonda M, Furness J W, et al. 2022 Phys. Rev. B 106 075422
[61] Sabatini R, Gorni T and De Gironcoli S 2013 Phys. Rev. B 87 041108

[1]	Observer-based prescribed-time time-varying output formation-containment control of heterogeneous multi-agent systems Haiyang Hu(胡海洋), Tao Li(李涛), Xiaowen Zhao(赵小文), Yuanmei Wang(王圆妹), Jialong Tian(田家龙), and Zijie Jiang(姜子杰). Chin. Phys. B, 2025, 34(10): 100503.
[2]	Two-dimensional Cr₂Cl₃S₃ Janus magnetic semiconductor with large magnetic exchange interaction and high-T_C Lei Fu(伏磊), Shasha Li(李沙沙), Xiangyan Bo(薄祥䶮), Sai Ma(马赛), Feng Li(李峰), and Yong Pu(普勇). Chin. Phys. B, 2024, 33(9): 096301.
[3]	Bipartite consensus problems of Lurie multi-agent systems over signed graphs: A contraction approach Xiaojiao Zhang(张晓娇) and Xiang Wu(吴祥). Chin. Phys. B, 2024, 33(7): 070204.
[4]	Fixed-time group consensus of second-order multi-agent systems based on event-triggered control Xiaoshuai Wu(武肖帅), Fenglan Sun(孙凤兰), Wei Zhu(朱伟), and Jürgen Kurths. Chin. Phys. B, 2023, 32(7): 070701.
[5]	Team-based fixed-time containment control for multi-agent systems with disturbances Xiao-Wen Zhao(赵小文), Jin-Yue Wang(王进月), Qiang Lai(赖强), and Yuan Liu(刘源). Chin. Phys. B, 2023, 32(12): 120502.
[6]	Memory-augmented adaptive flocking control for multi-agent systems subject to uncertain external disturbances Ximing Wang(王希铭), Jinsheng Sun(孙金生), Zhitao Li(李志韬), and Zixing Wu(吴梓杏). Chin. Phys. B, 2022, 31(2): 020203.
[7]	Fault-tolerant finite-time dynamical consensus of double-integrator multi-agent systems with partial agents subject to synchronous self-sensing function failure Zhi-Hai Wu(吴治海) and Lin-Bo Xie(谢林柏). Chin. Phys. B, 2022, 31(12): 128902.
[8]	Explosive synchronization in a mobile network in the presence of a positive feedback mechanism Dong-Jie Qian(钱冬杰). Chin. Phys. B, 2022, 31(1): 010503.
[9]	Pyramid scheme in stock market: A kind of financial market simulation Yong Shi(石勇), Bo Li(李博), and Guang-Le Du(杜光乐). Chin. Phys. B, 2021, 30(9): 098901.
[10]	Consensus problems on networks with free protocol Xiaodong Liu(柳晓东) and Lipo Mo(莫立坡). Chin. Phys. B, 2021, 30(7): 070701.
[11]	Distributed optimization for discrete-time multiagent systems with nonconvex control input constraints and switching topologies Xiao-Yu Shen(沈小宇), Shuai Su(宿帅), and Hai-Liang Hou(侯海良). Chin. Phys. B, 2021, 30(12): 120507.
[12]	Hybrid-triggered consensus for multi-agent systems with time-delays, uncertain switching topologies, and stochastic cyber-attacks Xia Chen(陈侠), Li-Yuan Yin(尹立远), Yong-Tai Liu(刘永泰), Hao Liu(刘皓). Chin. Phys. B, 2019, 28(9): 090701.
[13]	Group consensus of multi-agent systems subjected to cyber-attacks Hai-Yun Gao(高海云), Ai-Hua Hu(胡爱花), Wan-Qiang Shen(沈莞蔷), Zheng-Xian Jiang(江正仙). Chin. Phys. B, 2019, 28(6): 060501.
[14]	Successive lag cluster consensus on multi-agent systems via delay-dependent impulsive control Xiao-Fen Qiu(邱小芬), Yin-Xing Zhang(张银星), Ke-Zan Li(李科赞). Chin. Phys. B, 2019, 28(5): 050501.
[15]	Dynamics of an ultrasound contrast agent microbubble near spherical boundary in ultrasound field Ji-Wen Hu(胡继文), Lian-Mei Wang(王练妹), Sheng-You Qian(钱盛友), Wen-Yi Liu(刘文一), Ya-Tao Liu(刘亚涛), Wei-Rui Lei(雷卫瑞). Chin. Phys. B, 2019, 28(11): 114301.

No Suggested Reading articles found!

Viewed

Full text

Abstract

Cited

Metrics
Related Articles

VASPilot: MCP-facilitated multi-agent intelligence for autonomous VASP simulations

Cite this article:

Online attention