Role in the system
A2A provides a reference model for discovering remote agents and coordinating bounded tasks without exposing their internal memory or tools.
TECHNOLOGY
Agent2Agent is an open protocol for capability discovery, task coordination and exchange between independent AI agent systems.
EDITORIAL FRAME
A concise view of its scope, position, limitations and supporting sources.
A2A provides a reference model for discovering remote agents and coordinating bounded tasks without exposing their internal memory or tools.
A2A defines a shared language for independent agent applications. An agent can publish an Agent Card describing capabilities and interaction requirements, receive messages, manage long-running tasks and return artifacts without revealing its private prompt, memory or implementation.
A2A and MCP solve adjacent problems. MCP connects an AI application to context and tools; A2A coordinates work between agent applications. Neither protocol supplies authorization, trustworthy identity or safe autonomy by itself.
The protocol is open and governed through the Linux Foundation, but its interfaces are still evolving. Version 0.3.0 is a development milestone, not a guarantee of universal interoperability.
A2A is relevant to VASTE as a possible edge protocol for exchanging governed tasks between runtimes. VASTE would still own entity identity, permissions, provenance and state transitions.
DOCUMENTED RELATIONSHIPS
Each link names the relationship between two entries and why it matters.
Agent2Agent Protocol is an explicit member of the Knowledge Hub Fifth Wave collection.
A2A, Agent Interoperability and Governed Delegation documents Agent2Agent Protocol as one of its declared subjects.
Agent2Agent Protocol. 1.0.0. Electronic Artefacts, 2026-07-12. https://electronicartefacts.com/knowledge/technologies/agent2agent-protocol/
2 public links connect this page to nearby projects, concepts and references.