confighub-skills

Turn .env, properties, YAML, JSON, TOML, INI, or text into an AppConfig Unit and rendered ConfigMap via Upsert + render-configmap, or hash a hand-authored mutable ConfigMap with set-hash so linked workloads roll. Use for "use my .env", configMapGenerator-like config, envFrom, and a Deployment that never restarts on a ConfigMap change.

ConfigHub orientation and routing: vocabulary, the component model (Component / Variant / base / deployment), config-as-data, one resource per Unit, protection and merges, and delete/destroy gates. Use for tours, entity questions, topology, granularity, or broad help. Route concrete tasks to the dedicated skill.

Deprecated 0.4.x compatibility identity for prior cub-apply/apply/deploy requests. Grants no tools, performs no operation, and routes the unchanged intent to release-publish with its whole-Space scope and safety checks.

Change data inside an existing ConfigHub Unit, preferring a function over a hand-edit, and write it so it still works in every variant. Use for "update the image", "bump the replicas", "set the annotation", "apply defaults", a bulk edit, or "why did that change miss some variants?". Not for creating a Unit (kubernetes-resources).

Find, count, inspect, or audit Kubernetes config stored in ConfigHub — cub k8s get and cub resource list to browse resources, cub unit list for Units. Use for fleet sweeps and single-workload lookups; route "currently running/deployed" claims through verify-apply, because ConfigHub metadata alone is not runtime proof.

Onboard Helm charts or local Kustomize bases/overlays into Component/Variant/Release. Use for the cub-helm plugin (install/upgrade/template), multi-env variants, CRDs/hooks, preserving downstream edits, or rendered manifests + cub variant upload. Not for live-cluster or GitOps-CR adoption.

ConfigHub incident orchestration: bind the exact Release/runtime failure, choose rollback vs fix-forward, account for emergency cluster edits, and close with Release/controller/runtime proof. Use for outages, prod crashing, mitigation, and cleanup. Not planned releases.

Author a specific Kubernetes resource type as literal YAML in a ConfigHub Unit with best-practice defaults. Use for "create a StatefulSet", "add an Ingress", "set up NetworkPolicy", "I need a CronJob", "add RBAC for my app", "set up autoscaling", "add a PDB". Not for AppConfig-based ConfigMaps (use app-config).

Prepare Variant, ChangeOrder, or ChangeSet promotion, record and require approvals, and reconcile what a merge withheld. Use for "promote to staging", "roll forward to prod", "advance this change", "approve this change", "require approval", "which Units are behind?", partial promotion, or merge conflicts. Covers variant promote/approve, ChangeWorkflow, ChangeOrder, protection. Not rollback (rollback-revision).

Preview, publish, inspect, withdraw, or delete whole-Space OCI Releases. Use for apply/deploy/publish, pending/ChangeSet changes, preview, tagged selection, get/list, historical cancel questions, withdraw, or delete. Discloses EffectiveReleaseSet and provider races. Not data edits/rollback.

Roll back by moving one or more Unit heads to bound prior revisions via cub unit update --restore, then hand off to release-publish. Use for "roll back this change", "revert the last release", "undo the ChangeSet", or "restore to the last applied revision". Not for a clearer one-field forward fix.

Create or refresh a skill-examples Space with representative Kubernetes Units for exercising the other skills. Use for "set up the skill-examples space", "bootstrap the examples", "give me a Unit to tinker with", or "reset the examples". Not for real application Spaces (use confighub-core).

Prepare OCI Target, Space.ReleaseTargetID, and matching Unit.TargetID membership. Use for "set up a target", "publish this Space to OCI", "attach Units", or "do I need a worker?". Direct ConfigHub-provider delivery is historical and unsupported in the reviewed current profile.

Prepare or diagnose Trigger policy, ValidationErrors, and ValidationWarnings via platform Space + Filter + TriggerFilterID. Use for "block bad config", "enforce/warn", "require approval" (routed to ChangeWorkflow attestation prerequisites in promote-release), or "why is this Unit blocked?". Not one-off validation (cub-mutate).

Verify a published ConfigHub Space Release, then prove controller and cluster convergence read-only. Use after release-publish or for "did it deploy?", "is it live?", "did Argo pick it up?", and release close-out.

Explain and prepare worker setup. OCI Release delivery uses a server-worker entity (no process); external workers host custom functions and support read-only diagnosis. Use for "do I need a worker?", "host custom functions", or crashing workers. Direct ConfigHub-provider delivery is historical and unsupported in the reviewed current profile.