TY - JOUR
T1 - MCP agents in PanDA
AU - Nilsson, Paul
AU - Boudreau, Joseph
AU - Chowdhury, Tasnuva
AU - Feng, Shengyu
AU - Hoisie, Adolfy
AU - Khan, Raees
AU - Kim, Jaehyung
AU - Kilic, Ozgur O.
AU - Klasky, Scott
AU - Klimentov, Alexei
AU - Korchuganova, Tatiana
AU - Maeno, Tadashi
AU - Outschoorn, Verena Ingrid Martinez
AU - Park, David K.
AU - Podhorszki, Norbert
AU - Ren, Yihui
AU - Suter, Frédéric
AU - Vatsavai, Sairam Sri
AU - Wenaus, Torre
AU - Xue, Rui
AU - Yang, Wei
AU - Yang, Yiming
AU - Yoo, Shinjae
N1 - Publisher Copyright:
© 2026 IOP Publishing Ltd and Sissa Medialab. All rights, including for text and data mining, AI training, and similar technologies, are reserved. This article is available under the terms of the IOP-Standard License.
PY - 2026/7
Y1 - 2026/7
N2 - The PanDA workload management system, developed for large-scale distributed computing in high-energy physics, is being enhanced through the integration of AI-assisted operational tools built on the Model Context Protocol (MCP). This paper describes two complementary efforts. The first is PanDA MCP, a FastAPI-based interface layer that exposes PanDA REST APIs as standardized, self-describing MCP tools, bridging the synchronous PanDA backend with asynchronous AI clients. The second is Bamboo MCP, a modular plugin-based toolkit for AI-assisted operations, whose ATLAS plugin implements AskPanDA — a natural-language interface to the PanDA workload management system. Bamboo adopts a tool-first, evidence-driven architecture in which deterministic routing and structured data retrieval precede any LLM invocation. A key new capability enables natural-language queries against a live PanDA job database via an LLM-generated SQL pipeline protected by an AST-based security guard. A supervisor-managed suite of background agents maintains the local data stores on which these tools depend. The toolkit is experiment-agnostic by design, with plugins for ePIC, the Vera Rubin Observatory, and CGSim planned. A GPU-based testbed at Brookhaven National Laboratory supports co-development across the ATLAS and EIC communities.
AB - The PanDA workload management system, developed for large-scale distributed computing in high-energy physics, is being enhanced through the integration of AI-assisted operational tools built on the Model Context Protocol (MCP). This paper describes two complementary efforts. The first is PanDA MCP, a FastAPI-based interface layer that exposes PanDA REST APIs as standardized, self-describing MCP tools, bridging the synchronous PanDA backend with asynchronous AI clients. The second is Bamboo MCP, a modular plugin-based toolkit for AI-assisted operations, whose ATLAS plugin implements AskPanDA — a natural-language interface to the PanDA workload management system. Bamboo adopts a tool-first, evidence-driven architecture in which deterministic routing and structured data retrieval precede any LLM invocation. A key new capability enables natural-language queries against a live PanDA job database via an LLM-generated SQL pipeline protected by an AST-based security guard. A supervisor-managed suite of background agents maintains the local data stores on which these tools depend. The toolkit is experiment-agnostic by design, with plugins for ePIC, the Vera Rubin Observatory, and CGSim planned. A GPU-based testbed at Brookhaven National Laboratory supports co-development across the ATLAS and EIC communities.
KW - Computing (architecture, farms, GRID for recording, storage, archiving, and distribution of data)
KW - Software architectures (event data models, frameworks and databases)
UR - https://www.scopus.com/pages/publications/105046244376
U2 - 10.1088/1748-0221/21/07/C07030
DO - 10.1088/1748-0221/21/07/C07030
M3 - Article
AN - SCOPUS:105046244376
SN - 1748-0221
VL - 21
JO - Journal of Instrumentation
JF - Journal of Instrumentation
IS - 7
M1 - C07030
ER -