# MCP Registry: Connect AI agents with MCP tools — no shadow AI required
Build your enterprise directory for MCP tools. Register, discover, and govern MCP servers and tools so your AI agents connect only to trusted, approved resources — with full visibility and control.
## Close the registry gap before shadow AI takes hold
The Model Context Protocol (MCP) has simplified how agents connect to tools. But without a centralized, enterprise-approved directory, discovery is manual, security is fragmented, and shadow AI proliferates as agents connect to unverified, high-risk tools.
Kong MCP Registry solves this by giving your organization a single source of truth for every AI tool your agents can access — governed by the same security, identity, and observability you already use for APIs and events.
Govern where and how MCP servers are built, published, and consumed. Let developers build and ship fast while governance teams define the guardrails.
MCP should be an accelerant for agent development and AI adoption. Don’t let shadow AI concerns and manual MCP server discovery slow you down.
Reuse existing APIs and MCP tools for agentic workflows as opposed to creating redundant MCP servers for each new agentic use case.
## 01
## Dynamic discovery
Provide a centralized enterprise catalog where AI agents dynamically discover available MCP servers, API endpoints, and tools.
## 02
## Enterprise governance
Ensure only approved MCP servers and tools can be discovered and used, with clear ownership and metadata.
## 03
## Full observability
Centralize visibility into tool usage, health, and failures across all MCP servers.
