Agent Toolkits Community Session

The 22nd European Conference on Multi-Agent Systems (EUMAS 2025)

September 3, 2025 | Bucharest, Romania & online

The Agent Toolkits Community Session aims to bring together researchers and practitioners working in the domain of Engineering Multi-Agent Systems in an open and discussion-oriented setting — in which we welcome the sharing of research results, demonstrators, practical experiences, or position statements without the constraints of a formal peer-reviewing process.

This year's edition is hosted by EUMAS 2025 and will be held in a hybrid format, allowing for both in-person and online attendance. To register for online attendance, please fill out this form by August 27, 2025.

Organizers

• Katharine Beaumont, University College Dublin (IE)
• Andrei Ciortea, University of St.Gallen (CH) and Inria (FR)

Agenda

Click here to get the session's starting time according to your location.

Position Statements

Session 1: Agent Programming with LLMs

• Towards using LLMs to (co-)create agent systems [PDF]
  Michael Winikoff
• astra-langchain4j: Experiences Combining LLMs and Agent Programming [PDF]
  Rem Collier
• Agentic AI, Where Do We Go from Here? [PDF]
  Mehdi Dastani and I.S.W.B. Prasetya
• LLMs as Agents, LLMs at the Service of Agents, or Agents at the Service of LLMs? [PDF]
  Gianluca Aguzzi, Giovanni Ciatto, Angelo Ferrando, Andrea Gatti, and Viviana Mascardi

Session 2: Design Dimensions, Tool Use, and Communities

• Using (Traditional) Multi-Agent Principles to Suggest Improvements for Agentic AI Platforms [PDF]
  Guillaume Muller, Somsakun Maneerat, Héloïse Pajot, Chloé Petridis, Richard Niamke
• Empowering the Concept of Tool in Agentic AI: Co-evolving AI Agents and Environments with A&A and Hypermedia MAS [PDF]
  Alessandro Ricci and Andrei Ciortea
• Experience-based Agent Communities in support of Smart Environment Hypermedia MAS [PDF]
  Alexandru Sorici and Andrei Olaru
• Social Simulation and Multi-Agent Systems: A False Dichotomy in the Age of Generative Agents [PDF]
  Vivek Nallur

Session 3: Interaction, Interoperability, and Governance

• Multiagent Interaction Models for Agentic AI [PDF]
  Matteo Baldoni, Cristina Baroglio, Amit K. Chopra, Elisa Marengo, Roberto Micalizio, and Stefano Tedeschi
• Coral Protocol: Applying Agent-Oriented Software Engineering Principles to Generative-AI Ecosystems [PDF]
  Önder Gürcan
• Language Agents May Enable Interoperability in Heterogeneous Multi-Agent Systems [PDF]
  Jérémy Lemée and Simon Mayer
• Why We Need Event Data Models for Agents and Multi-Agent Systems [PDF]
  Timotheus Kampik