MadAgents — Draconitic Research

Abstract

We uncover an effective and communicative set of agents working with MadGraph. Agentic installation, learning-by-doing training, and user support provide easy access to state-of-the-art simulations and accelerate LHC research. We show in detail how MadAgents interact with inexperienced and advanced users, support a range of simulation tasks, and analyze the results. In a second step, we illustrate how MadAgents automatize event generation and run an autonomous simulation campaign, starting from a pdf file of a paper. We also present successive MadAgents updates, including a Claude Code implementation with a self-improvement loop.

Cite

BibTeX RIS

BibTeX

@article{plehn2026madagents,
  title = {{MadAgents}},
  author = {Tilman Plehn and Daniel Schiller and Nikita Schmal},
  journal = {SciPost Physics Codebases},
  year = {2026},
  month = nov,
  doi = {10.21468/scipostphyscodeb.86},
  url = {https://doi.org/10.21468/scipostphyscodeb.86},
  note = {CC BY},
}

RIS

TY  - JOUR
TI  - MadAgents
AU  - Tilman Plehn
AU  - Daniel Schiller
AU  - Nikita Schmal
PY  - 2026
DA  - 2026/11/05
JO  - SciPost Physics Codebases
DO  - 10.21468/scipostphyscodeb.86
UR  - https://doi.org/10.21468/scipostphyscodeb.86
AB  - We uncover an effective and communicative set of agents working with MadGraph. Agentic installation, learning-by-doing training, and user support provide easy access to state-of-the-art simulations and accelerate LHC research. We show in detail how MadAgents interact with inexperienced and advanced users, support a range of simulation tasks, and analyze the results. In a second step, we illustrate how MadAgents automatize event generation and run an autonomous simulation campaign, starting from a pdf file of a paper. We also present successive MadAgents updates, including a Claude Code implementation with a self-improvement loop.
SN  - 2949-804X
ER  -