Skip to content
Deployment
Command

/migrate-k8s

Migrate Kubernetes manifests, Docker Compose projects, or Helm charts to Control Plane

From plugin
ai-plugin
102 skills2 agents2 commands
Install
$ npx -y skills add controlplane-com/ai-plugin --agent claude-code

How it fires

How this command gets triggered: by you, by Claude, or both.

  • Fires itselfClaude auto-loads it when your prompt matches the work.
  • You can call itInvoke it directly when you want it.
  • Slash command/migrate-k8s

Context preview

What this command does when you run it.

Migrate Kubernetes manifests, Docker Compose projects, or Helm charts to Control Plane

Command definition

migrate-k8s.md
name: migrate-k8s
description: Migrate Kubernetes manifests, Docker Compose projects, or Helm charts to Control Plane
argument-hint: "[path-to-manifest-or-directory] [--from k8s|compose|helm]"

Migrate the Kubernetes, Docker Compose, or Helm source the user pointed to onto Control Plane.

Use the **cpln-k8s-migrator** agent — it loads the `migration-patterns` skill (the canonical conversion reference) and carries the migration end to end:

1. Identify the source and pick the converter (CLI-only, no MCP equivalent): `cpln convert` for Kubernetes manifests (pipe `helm template` output in for a Helm chart), `cpln stack` for Docker Compose, or `cpln helm` to deploy a chart that already renders Control Plane resources. 2. Convert to a file and review it — diff the source against the output for what the converter changed or silently dropped (pinned scaling, `envFrom`, init containers, NetworkPolicy/RBAC, minimal sizing). 3. Provision in dependency order, MCP-first via the typed `create_*` tools (GVC, then identities, volumesets, workloads, domains; secrets are created by the user first — draft the manifests for them to fill and apply); fall back to `cpln apply -f` when MCP is unavailable or in CI/CD. 4. Verify every workload is ready across its locations and report the canonical endpoint — never a constructed URL.

If the org or target GVC isn't named, ask — never guess.

Ships withai-plugin

Run containerized workloads across AWS, GCP, Azure, OCI, and your own hardware under one API.

Get the whole plugin, auto-invoked
Stats
10
Stars
0
Views
0
Forks
Active
Maintenance
Shell
Language
MIT
License
2d ago
Last commit
3mo ago
Created

Repo: controlplane-com/ai-plugin