Agent Registry data model

Agent Registry categorizes agentic components of your ecosystem into four main resource types that you can manage and discover:

  • Agents: Autonomous AI actors that carry out specific AI workflows. An agent understands its workflow capabilities by possessing specific skills. In Agent Registry, agent skills are represented as declarative metadata (A2A skills) or as a modular, executable package (Skills).
  • MCP servers: Providers of standardized data resources and tools. Tools are deterministic functions exposed by an MCP server that agents can invoke to perform specific actions.
  • Endpoints: Target URLs, typically REST APIs, accessed by an agent. By abstracting these destinations into manageable resources, Agent Registry lets you centrally govern which external APIs your agents can access.
  • Skills: Standalone, executable packages of instructions, code, and metadata governed directly within Agent Registry. Standalone skills are distinct from A2A skills, which are declarative metadata block manifests declared inline within an agent's Agent Card.

For more information about these and other concepts that Agent Registry uses, see Key concepts.

API resources

The Agent Registry API separates read and write operations for agents, endpoints, and MCP servers:

  • Writable API resource: You manage these agentic components by manually creating, updating, or deleting a Service resource in the registry.
  • Read-only API resource: To query, list, or search for these agentic components, you request the read-only Agent, McpServer, or Endpoint resource to the Agent Registry API.

Standalone skills don't use the Service projection model. Instead, you manage, version-control, and list skills directly using the Skill, SkillRevision, and Publisher API resources.

To get started, see Set up Agent Registry. For more information about the Agent Registry API, see the API reference documentation.

Register agentic components in your registry

Agent Registry maintains your catalog of agents and MCP servers through two registration mechanisms: automatic and manual.

Agent Registry automatically registers agents and servers deployed on supported Google Cloud runtimes. Register and manage your agents, MCP servers, and endpoints by enabling the Agent Registry API. Google-created skills are also visible immediately when you set up Agent Registry.

Additionally, you can use the Service resource to create, update, and delete entries manually. Then, you query the read-only Agent, McpServer, or Endpoint resources to get, list, and search for your registered agents, servers, and endpoints, respectively. For manual registration and search of skills, use the Skill resource.

Registration mechanism Description Registered component Supported Google Cloud products
Automatic registration You can configure and use Agent Registry to automatically discover and ingest specific resources deployed on supported Google Cloud runtimes. For agents implementing the open Agent2Agent (A2A) protocol, the registry also automatically extracts their A2A skills and capabilities from their Agent Card. Agents
MCP servers Official Google and Google Cloud remote MCP servers
Skills Google's public skills are globally discoverable from all projects
Manual registration For external resources, custom agentic components on unsupported runtimes, or standard REST agents that don't implement the A2A protocol, you must manually register them by creating a Service resource in the registry. For standalone skills, you use the Skill resource. Agents Remote agents
MCP servers Custom agentic components on unsupported runtimes or external MCP servers
Skills Your own standalone skills

What's next