Description

The rapid emergence of agentic AI systems, capable of autonomous reasoning, planning, and decision-making, is creating a new generation of distributed ecosystems in which agents continuously communicate, coordinate, negotiate, and collaborate with each other. As organizations increasingly deploy large populations of autonomous agents across cloud infrastructures, communication networks, edge environments, and digital services, agent-to-agent communication is becoming a critical research challenge.  MAGIC focuses on the communication and networking foundations required to support large-scale agent ecosystems. Future agentic environments will require agents to discover one another, exchange information, advertise capabilities, negotiate tasks, coordinate actions, establish trust, and collaborate across heterogeneous platforms and administrative domains through the creation of agentic overlay networks. Achieving this vision raises fundamental questions about communication protocols, interoperability standards, service discovery mechanisms, communication efficiency, scalability, reliability, and security. 

Beyond individual interactions, the workshop explores how networks of agents can be managed and governed. As agent populations grow, new challenges emerge around communication orchestration, coordination policies, trust establishment, identity management, observability, monitoring, accountability, and communication control.  Attention is given to emerging paradigms such as agent-to-agent (A2A) communication, multi-agent systems, federated agent ecosystems, decentralized coordination, and human-agent-agent collaboration. Together, these challenges define a new research agenda at the intersection of communication networks, distributed systems, and agentic AI. MAGIC aims to bring together researchers and practitioners working on the protocols, architectures, and management frameworks that will enable scalable, secure, interoperable, and trustworthy communication among autonomous agents. 

 Topics include, but are not limited to: 

 • Agent-to-Agent (A2A) communication protocols and architectures 

• Architectures and frameworks for agent ecosystems 

• Multi-agent coordination, negotiation, and collaboration 

• Interoperability across heterogeneous agent frameworks 

• Agent discovery, registration, and capability advertisement 

• Semantic communication and knowledge exchange among autonomous agents 

• Trust, identity, authentication, and security for communicating agents 

• Learning-based coordination and decision-making in multi-agent systems 

• Agent orchestration, governance, and lifecycle management 

• Tool use, service discovery, and function invocation among agents 

• Distributed and federated agent ecosystems 

• Agent observability, monitoring, and communication analytics 

• Digital twins and simulation platforms for agent ecosystems 

• Performance evaluation and benchmarking of agent communication systems 

 Paper Submission via EDAS:  https://edas.info/N35575