Developer Preview features

This section lists Developer Preview features in Red Hat Developer Hub 2.1.

Developer Preview features are not supported by Red Hat in any way and are not functionally complete or production-ready. Do not use Developer Preview features for production or business-critical workloads. Developer Preview features provide early access to functionality in advance of possible inclusion in a Red Hat product offering. Customers can use these features to test functionality and provide feedback during the development process. Developer Preview features might not have any documentation, are subject to change or removal at any time, and have received limited testing. Red Hat might provide ways to submit feedback on Developer Preview features without an associated SLA. For more information about the support scope of Red Hat Developer Preview features, see Developer Preview Support Scope.

Entity provider for MCP Registry services (Developer Preview)

This update introduces a catalog plugin module, known as an entity provider, that provides Model Context Protocol (MCP) server API entities in the Red Hat Developer Hub catalog from a target MCP Registry service. The entity provider connects to any MCP Registry by using the official MCP Registry specification (io.modelcontextprotocol.registry).

Updates to the OpenShift AI connector (Developer Preview)

This update to the OpenShift AI connector includes the following changes:

  • The sidecar containers are removed, and their functions are replicated inside a dynamic plugin.
  • A new AiModelServerAPI catalog entity kind is introduced, which replaces the use of the Component, Resource, and API entity kinds.
  • Support for the Kubeflow Model Registry from OpenShift AI is removed.
  • Support for KServe InferenceService resources and the Kubeflow Model Catalog from OpenShift AI is maintained.
  • Support for KServe LLMInferenceService resources is added.

Red Hat Developer Hub on ARM64 clusters (Developer Preview)

Starting with Red Hat Developer Hub (Developer Hub) 2.1, you can run Developer Hub on 64-bit Arm (ARM64, also known as aarch64) clusters as a Developer Preview by using the community container images at quay.io/rhdh-community. You can deploy Developer Hub on an ARM64 Kubernetes cluster, such as Red Hat OpenShift Container Platform on 64-bit Arm or a cloud provider that uses 64-bit Arm compute nodes. Running Developer Hub on ARM64 lets you host your developer portal on Arm-based infrastructure. Red Hat validates both the Developer Hub Helm chart and the Developer Hub Operator on ARM64 with the community images.

Developer Preview validation confirms that pods reach the Ready state, the Developer Hub portal loads, and basic navigation works. This Developer Preview does not include productized multi-architecture images. Red Hat does not validate the Intelligent Assistant (formerly Developer Lightspeed for RHDH) on ARM64. If you deploy Developer Hub on an ARM64 cluster, disable the Intelligent Assistant.

For more information, see Deploying Red Hat Developer Hub on ARM64 clusters.

AI-assisted software template authoring (Developer Preview)

A community skill for AI coding agents helps platform engineers create Red Hat Developer Hub software templates that follow project conventions. You can use it to scaffold a template from scratch or from existing code, validate templates locally, and discover scaffolder actions.

Red Hat Developer Hub introduces the AiResource entity kind (Developer Preview)

This update introduces the AiResource entity kind in Red Hat Developer Hub as a Developer Preview. You can register Git repositories or Open Container Initiative (OCI) image references that contain AI skills and rules, discover them through catalog search and filtering, and register them through the Intelligent Assistant (formerly Developer Lightspeed for RHDH) when the Model Context Protocol (MCP) is configured.

Red Hat Developer Hub CLI for querying the organizational knowledge graph (Developer Preview)

This update introduces intent-based commands in the Red Hat Developer Hub CLI (rhdh-cli) as a Developer Preview. You can query and manage the Red Hat Developer Hub (RHDH) organizational knowledge graph directly from the command line or an AI coding agent to ground architecture decisions in your organization’s actual services, APIs, teams, and standards.

With rhdh-cli, you can:

  • List, get, validate, register, and unregister catalog entities, such as components, APIs, systems, and resources.
  • Retrieve full API specifications (OpenAPI, AsyncAPI, GraphQL, and gRPC) to understand service interfaces before you integrate with them.
  • Search across catalog entities, TechDocs, and templates to discover existing assets.
  • Discover, validate, dry-run, and execute software templates to create projects that comply with your organization’s standards.

Human-readable output supports interactive use, and structured JSON output (--output=json) helps developers and AI coding agents use organizational context and automate workflows. Every command supports --help to expose the full command surface, so that AI agents can discover and operate the CLI without prior setup.

To use rhdh-cli with RHDH, enable the @backstage/plugin-auth front-end plugin and configure the OAuth2 endpoints.

Agent Skills support for the Intelligent Assistant (Developer Preview)

This update introduces Agent Skills support for the Red Hat Developer Hub Intelligent Assistant (formerly Developer Lightspeed for RHDH) as a Developer Preview. Platform engineers can configure skills with organization-specific instructions and supporting reference material. The Intelligent Assistant selects relevant skills when it responds to prompts, so that answers align with your organization’s coding standards, troubleshooting procedures, and approved workflows.

The DevHubPluginCatalog custom resource manages plugin catalogs as Kubernetes resources (Developer Preview)

You can manage plugin catalogs as namespace-scoped Kubernetes resources by using the DevHubPluginCatalog custom resource (CR). This CR lets you combine multiple catalogs, such as Red Hat Developer Hub official catalogs and custom catalogs, validate catalogs before deployment, and inspect catalogs by using standard kubectl commands. The DevHubPluginCatalog controller fetches catalogs from Open Container Initiative (OCI) images and merges them into a unified default-config ConfigMap.

apiVersion: rhdh.redhat.com/v1alpha5
kind: DevHubPluginCatalog
metadata:
  name: rhdh-catalog
  namespace: rhdh-operator
spec:
  source:
    ref: oci://quay.io/rhdh/plugin-catalog-index:next

This update provides the following benefits:

  • Multi-catalog support: Deploy Red Hat Developer Hub official catalogs and custom catalogs at the same time.
  • Early validation: Catalog fetch errors appear in the Ready condition before the Backstage deployment.
  • Standard Kubernetes lifecycle: Use kubectl to inspect, manage, and troubleshoot catalogs.

This feature requires Operator processing mode (OPERATOR_DP_PROCESSING=true).

Plugin configuration processing moves to Operator reconciliation (Developer Preview)

Plugin configuration processing moves to Operator reconciliation, which enables early validation and faster deployments. This update provides the following improvements:

  • Up to 10 times faster init container.
  • Three status tiers: Config, Deployed, and Runtime conditions.
  • The plugin list is visible in status.plugins.
  • Configuration errors are caught during reconciliation instead of at pod startup.

To enable Operator processing mode, set OPERATOR_DP_PROCESSING=true on the Operator deployment. The default is false.

When Operator processing mode is enabled, you can also automatically rewrite all oci:// plugin and catalog image references to use internal or mirrored registries for disconnected environments. This uses the Red Hat Developer Hub on OpenShift Container Platform ImageDigestMirrorSet (IDMS) and ImageTagMirrorSet (ITMS) compatible format with a most-specific-match rule:

apiVersion: v1
kind: ConfigMap
metadata:
  name: plugins-mirror-config
  namespace: rhdh-operator
data:
  mirrors.yaml: |
    imageMirrors:
      - source: quay.io/rhdh
        mirrors:
          - internal-registry.company.com/rhdh

OpenShift AI connector adds model discovery and agent support (Developer Preview)

This update to the OpenShift AI connector adds the following capabilities:

  • An entity provider that queries the models endpoint of a configured Open GenAI Stack (OGX) endpoint provided with OpenShift AI, and stores the returned models as AiModelServerAPI entities.
  • A new agent type for the AiResource entity kind.
  • Support for declaring OpenAI Agent SDK-compatible agents in app-config.yaml. These agents are stored in the catalog as AiResource entities of the agent type.

Automatic rewriting of image references for disconnected environments (Developer Preview)

When Operator processing mode is enabled (OPERATOR_DP_PROCESSING=true), you can automatically rewrite all oci:// plugin and catalog image references to use internal or mirrored registries for disconnected environments. This uses the Red Hat OpenShift Container Platform ImageDigestMirrorSet (IDMS) and ImageTagMirrorSet (ITMS) compatible format with a most-specific-match rule.

apiVersion: v1
kind: ConfigMap
metadata:
  name: plugins-mirror-config
  namespace: rhdh-operator
data:
  mirrors.yaml: |
    imageMirrors:
      - source: quay.io/rhdh
        mirrors:
          - internal-registry.company.com/rhdh