MCP registries in Catalog (tech preview)

An MCP Registry is a centralized publishing and discovery endpoint for MCP servers within your organization.

As organizations experiment with AI agents, MCP servers are often embedded directly into agent code, stored in local configuration files, or scattered across repositories. Over time, this can lead to MCP sprawl, making it difficult to understand:

• What MCP servers exist
• Which agents should use which servers
• How those servers should be governed

MCP Registries extend Konnect Catalog to provide a structured, standards-based way to register and manage MCP servers, helping Platform Teams maintain visibility and control as AI adoption scales.

This feature is built on top of the MCP Registry API specification defined by Anthropic’s open source project. The specification allows flexibility in how registries are structured, so you can model them according to your organization’s governance needs.

MCP Registries in Catalog are available in tech preview. Contact Kong to have this feature enabled for your organization.

Once enabled, you can access MCP Registries by doing the following:

1. In the Konnect sidebar, click Catalog.

2. Click the MCP registries tab.

   If you’re using Catalog Classic, click MCP Registries in the sidebar instead.

1. In the Konnect sidebar, click Catalog.

2. From the New dropdown menu, select “MCP registry”.

   If you’re using Catalog Classic, click MCP Registries in the sidebar instead.

3. In the Display Name field, enter a name for your registry, for example Production Registry.
4. In the Name field, enter a unique identifier for the registry, for example production-registry. This must be lowercase and alphanumeric, and can include hyphens.
5. (Optional) In the Description field, describe the purpose of this registry.
6. Click Create.

An MCP server represents an agent-facing service definition. It describes:

• What the server does
• Its version
• How agents can connect to it

1. In the Konnect sidebar, click Catalog.
2. Click the MCP registries tab.
3. Click your MCP registry.
4. From your MCP Registry’s detail page, click New MCP Server.
5. (Optional) In the Title field, enter a short, descriptive name for the server, for example Filesystem Server.
6. In the Name field, enter a unique identifier in reverse-DNS format, for example io.example/filesystem. This must contain exactly one forward slash.
7. In the Version field, enter a version number using semantic versioning, for example 1.0.0.
8. In the Description field, enter a description of what the server does.
9. (Optional) Click Show optional fields, and do the following:
   1. (Optional) In the Schema URI field, enter a URI, for example https://example.com/schemas/server.json.
   2. (Optional) In the GitHub repository URL field, enter a URL, for example https://github.com/owner/repo.
   3. (Optional) In the Website URL field, enter a URL, for example https://example.com.
   4. (Optional) In the Metadata field, enter a JSON object, for example {"license": "MIT", "author": "Example Corp", "certified": true}.
10. (Optional) To describe a run-it-yourself distribution option, such as an npm or PyPI package, click Add Package and do the following:
    1. From the Transport type dropdown menu, select an option. If you select “Streamable-HTTP” or “SSE”, also enter the transport URL.
    2. From the Registry type dropdown menu, select how you want users to download packages.
    3. In the Registry base URL field, enter https://registry.npmjs.org.
    4. In the Package identifier field, enter the base URL of the package registry, for example @modelcontextprotocol/mcp-server.
    5. (Optional) In the Package version field, enter a version, for example 1.0.0.
    6. Click Save.
11. (Optional) To add a remote to describe a hosted MCP server endpoint that agents can connect to over the network, click Add Remote and do the following:
    1. From the Transport type dropdown menu, select “Streamable-HTTP” or “SSE”.
    2. In the Remote URL field, enter the MCP server’s endpoint, for example https://example.com/mcp.
    3. (Optional) In the Headers field, enter headers as a JSON array of objects, for example, to pass an authorization token:

       [
         {
        "name": "Authorization",
        "value": "Bearer ${token}"
         }
       ]

       Copied!
    4. Click Save.
12. Click Publish.

MCP servers are defined independently from a single deployment. They may include multiple delivery paths.

Packages

Packages describe run-it-yourself distribution options, such as:

• npm packages
• PyPI packages
• OCI artifacts

These allow teams to install and run the MCP server within their own agent environments.

Remotes

Remotes describe hosted MCP server endpoints that agents can connect to over the network, such as:

• streamable-http
• sse

An MCP server can define multiple packages and multiple remotes simultaneously. This allows organizations to support different runtime environments without duplicating server definitions.

Registry endpoints assume a Konnect authentication context. Only authenticated clients with appropriate access tokens can query a registry.

We are continuing to evolve MCP Registries alongside the broader MCP ecosystem.

Planned enhancements include:

• Linking Catalog MCP servers to MCP servers created in AI Gateway 2.0
• Additional governance and lifecycle controls

Because the MCP specification is still evolving, we are committed to iterating in partnership with customers who have already begun developing MCP servers and experimenting with agent-based workflows.