AWS Launches Agent Registry for Enterprise AI Agent Discovery and Governance
Amazon Web Services has announced the AWS Agent Registry, a centralized platform for discovering, sharing, and reusing AI agents, tools, and agent skills across enterprises. The announcement comes amid a proliferation of AI agent registries from major cloud providers.
The Problem
As enterprises deploy growing numbers of AI agents, they face a visibility crisis: teams don't know what agents exist, who owns them, or what they do. This leads to redundant development, inconsistent governance, and security risks from shadow AI tools.
AWS Agent Registry Features
- Central discovery hub: Single place to find all AI agents across an organization
- Metadata management: Ownership, capabilities, protocol details, invocation instructions
- Standards support: MCP (Model Context Protocol), A2A (Agent-to-Agent), custom schemas
- Automatic indexing: Agents built with AgentCore, Amazon Quick Suite, and Kiro auto-indexed
- Cross-platform: Works with agents hosted on AgentCore, other clouds, or on-premises
- Query interfaces: AWS console, SDKs, APIs, MCP-compatible clients like Claude Code
The Registry Arms Race
AWS is not alone in this space:
| Provider | Registry | Status |
|---|---|---|
| AWS | Agent Registry (via AgentCore) | Preview |
| Microsoft | Entra Agent Registry + Azure Agent Registry | Available |
| Google Cloud | Agent Registry (IAM) | Available |
| Open Source | ACP (Agent Client Protocol) Registry | Available |
Availability
Preview in five regions: US East (N. Virginia), US West (Oregon), Asia Pacific (Sydney), Asia Pacific (Tokyo), and Europe (Ireland).
Significance
The simultaneous launch of agent registries by all major cloud providers signals that AI agents are maturing from experimental tools to enterprise infrastructure that requires the same governance, discovery, and lifecycle management as traditional microservices and APIs.