Original article excerpt
Server-side extracted preview paragraphs from the original source.
In this technical collaboration between AWS and the authors, we present a pragmatic solution: agentic overlays. Agentic overlays are thin wrapper layers that transform traditional REST-based services into agents capable of participating in A2A interactions. They also expose REST APIs as tools compatible with the Model Context Protocol (MCP). Together, they let enterprises add A2A capabilities to existing REST services without rewriting business logic, without duplicating code, and without running parallel infrastructures. This reduces agent sprawl in the infrastructure by reusing existing services as agents. We provide reference architectures and sample code that show how to build agentic overlays.
The opinions expressed in this post are the authors’ views and not those of Cisco.
Enterprise architectures have long been centered on REST APIs and microservices. These systems are stable, well-tested, and deeply embedded in production environments. They weren’t designed for Agent-to-Agent (A2A) communication, the emerging standard for autonomous agents that collaborate, reason, and coordinate through structured messaging. That worked in the absence of a common agent protocol, but it means many existing agents now sit outside the emerging A2A framework. The challenge today is no longer only adding A2A to traditional services. You also need to bring these REST-based agents into a standardized agent-to-agent world.
In this technical collaboration between AWS and the authors, we present a pragmatic solution: agentic overlays. Agentic overlays are thin wrapper layers that transform traditional REST-based services into agents capable of participating in A2A interactions. They also expose REST APIs as tools compatible with the Model Context Protocol (MCP). Together, they let enterprises add A2A capabilities to existing REST services without rewriting business logic, without duplicating code, and without running parallel infrastructures. This reduces agent sprawl in the infrastructure by reusing existing services as agents. We provide reference architectures and sample code that show how to build agentic overlays.
REST APIs are designed for deterministic, client-server integration. A client calls a well-defined endpoint, passes parameters, and receives a predictable response, typically in a stateless request-response flow governed by HTTP semantics. This makes REST excellent for exposing business capabilities (such as create, read, update, and delete) with clear contracts, strong compatibility, and operational simplicity.
A2A is designed for interoperability between autonomous agents. Agents discover one another through metadata (such as an agent card), negotiate capabilities, and exchange structured messages (often through JSON-RPC) to coordinate multi-step tasks. Where REST optimizes for stable service interfaces and direct execution, A2A optimizes for reasoning-driven coordination, task-oriented messaging, and agent collaboration. The result is systems that can plan, delegate, and compose actions across multiple services rather than invoking isolated endpoints.
REST APIs and agentic systems are based on orthogonal paradigms, which makes it hard for enterprises to move existing services into standardized agentic communication. Yet enterprises need to use both without a major overhaul. Although newer agent communication through A2A introduces coordination models for enterprise systems, adoption has been slowed by the need to deploy and operate agentic infrastructures alongside existing enterprise systems. This parallel operation increases operational complexity and cost, creating barriers to adopting AI effectively.