The Agentic Web is emerging as billions of AI agents discover, communicate, and coordinate across the open Web, marking a shift toward systems that no longer operate as isolated models but as Web-integrated entities. This workshop unifies architectures and standards with the foundations of interoperable ecosystems, convening researchers and practitioners to chart a shared agenda.
Recent efforts suggest growing activity around interoperable agent ecosystems, including the
NANDA Index & Verified AgentFacts
and the
NEST sandbox/testbed,
alongside industry-led protocol work such as Google’s
A2A
and
AP2,
Cisco-backed
AGNTCY,
and interface standards like
MCP.
Taken together, these parallel tracks make it timely to convene researchers and practitioners to compare approaches and align on open, interoperable foundations for the Agentic Web.
We focus on web-native building blocks: agent registries and resolution, identity and credentials (DIDs/VCs), authorization (OAuth 2.0), discovery (DNS-SD), and federation patterns (e.g., ActivityPub), and their interfaces with application-level protocols (A2A, MCP, OpenAPI/REST/GraphQL/gRPC) and corresponding architectural models.
Beyond wiring, we examine economic mechanisms (reputation under adversarial conditions, knowledge pricing, transaction protocols) and societal coordination (governance, accountability, evaluation at population scale). Topics include capability representation and matching, cross-protocol bridges, privacy and provenance, workflow orchestration, and reliable tool use over heterogeneous data and services spanning cloud, edge/IoT, and enterprise environments.
The program blends keynote, invited talks, papers, and a panel to surface design principles, pitfalls, and testbed practices. By aligning standards, infrastructure, and incentives, the workshop seeks to ensure the Agentic Web remains open, trustworthy, and sustainable.