| name | rhoai-connected-applications |
| metadata | {"author":"rhoai3-demo","version":"1.0.0","platform-family":"rhoai","platform-baseline":"repo","ocp-baseline":"repo","skill-group":"RHOAI Platform"} |
| description | Use when documenting, reviewing, or operating Red Hat OpenShift AI connected application workflows from the Working with connected applications guide: viewing applications from Applications -> Explore, enabling SaaS-based connected applications, understanding Operator-installed application boundaries, finding API endpoints on Enabled tiles, removing disabled application tiles from the dashboard, starting and updating the Start basic workbench connected application, and routing basic workbench JupyterLab notebook, Git, and package workflows. Do NOT use for administrator dashboard application tile authoring or OdhApplication/OdhDashboardConfig manifests (use rhoai-dashboard-applications), global user/group access management (use rhoai-users-groups-access), project workbench lifecycle (use rhoai-project-workflows), general JupyterLab/code-server IDE workflows (use rhoai-data-science-ide-workflows), basic workbench administrator controls (use rhoai-basic-workbenches), or live cluster changes without the OpenShift safety guard.
|
RHOAI Connected Applications
Use this skill for user-facing connected application workflows in the
OpenShift AI dashboard on the active product baseline in
docs/PLATFORM_BASELINE.md.
Source Grounding
Read references/source-capture.md before using product workflow details.
Official Red Hat documentation is product authority. This skill adapts the
official Working with connected applications guide to this repo's demo
workflow and documentation model.
Scope
This skill covers:
- viewing available connected applications from Applications -> Explore
- identifying when an Enable button is available on a tile
- enabling SaaS-based connected applications from the dashboard
- understanding on-cluster and Operator-installed application boundaries
- finding API endpoints and resource links on enabled application tiles
- removing tiles for applications that an administrator has already disabled
- using the Start basic workbench connected application
- selecting basic workbench images, versions, container size, accelerator
count, and environment variables
- restarting a basic workbench to update settings
- routing basic workbench JupyterLab notebook, Git, and package workflows to
the IDE workflow skill
Use other skills for adjacent work:
rhoai-dashboard-applications for administrator-side dashboard application
tile resources, connected application disablement, and OdhApplication or
OdhDashboardConfig review
rhoai-users-groups-access for OpenShift AI user/admin group membership
and access permission errors
rhoai-project-workflows for preferred project workbench creation,
connections, storage, and project lifecycle
rhoai-data-science-ide-workflows for JupyterLab notebook, Git, package,
and code-server workflows
rhoai-basic-workbenches for administrator-side basic workbench management,
idle timeout, pod tolerations, and troubleshooting
rhoai-nvidia-gpu-accelerators for accelerator enablement and GPU workbench
support posture
Demo Policy
For this repo:
- Prefer project-based workbenches for the main demo architecture because they
organize data science work, connections, storage, models, and pipelines in a
project boundary.
- Use Start basic workbench only for lightweight Jupyter access, imported
notebooks with no project dependencies, or administrator-approved simple
demos.
- Do not treat dashboard connected application tiles as proof that the backing
application, Operator, endpoint, or credentials are installed and working.
- Do not add or remove dashboard tiles in GitOps through this skill; use
rhoai-dashboard-applications for administrator-side tile manifests.
- Treat SaaS service keys and application credentials as secrets. Never commit
them in READMEs, notebooks, scripts, or examples.
- Use endpoint URLs from Enabled tiles only after verifying the application
actually exposes an endpoint; some applications provide resources such as
notebook images rather than endpoint links.
- Respect namespace requirements documented on application tiles, especially
for independent software vendor applications.
- Mark software-catalog or third-party Operator deployments as potentially
outside full Red Hat support unless an official support statement confirms
otherwise.
- Use accelerator selections only when accelerator support is enabled and the
selected workbench image supports the requested accelerator type.
Workflow
- Confirm the active baseline in
docs/PLATFORM_BASELINE.md.
- Read
references/source-capture.md and
references/official-doc-extraction.md.
- Decide whether the task is:
- connected application discovery
- SaaS connected application enablement
- disabled tile removal
- API endpoint or resource link review
- Start basic workbench access or settings update
- JupyterLab handoff to IDE workflow guidance
- Use
examples/connected-application-patterns.md for focused review
patterns.
- For live cluster work, follow the OpenShift safety guard in
AGENTS.md.
- Validate with
references/validation-checklist.md.
References
references/source-capture.md
references/official-doc-extraction.md
references/validation-checklist.md
examples/connected-application-patterns.md