Lists multiple specific concrete actions: creates Units pre-bound to a cluster Target, runs `cub unit import` with `--where-resource` filters to pull live manifests, then hands off to `config-as-data` doctrine for ongoing management.

Clearly answers both 'what' (creates Units, runs cub unit import with filters, pulls live manifests) and 'when' (explicit 'Use when' clause at the start with detailed trigger scenarios). Also includes explicit 'Do not load for' exclusions which further clarify when to use it.

Excellent coverage of natural phrases users would say: 'adopt these existing resources into ConfigHub', 'reverse engineer my namespace into Units', 'import from the cluster', 'cub unit import', 'bring the running state into ConfigHub', 'we kubectl-apply'd everything and want to migrate'. These are realistic, varied phrasings.

Exceptionally distinctive — explicitly delineates boundaries by listing four specific skills to use instead for Helm charts, ArgoCD, Flux, and new YAML authoring. The niche of importing already-running Kubernetes resources without a GitOps tool is very clearly carved out.

The skill is reasonably efficient but includes some unnecessary framing (e.g., the 'Positioning: this is an onboarding tool' section restates what the description already conveys, and some explanatory prose like 'This skill targets users with existing cluster state' could be trimmed). The 'Do not load for' section partially duplicates the frontmatter description. However, most content earns its place — the scoping examples, workflow steps, and stop conditions are all valuable.

The skill provides fully executable CLI commands throughout — `cub unit create`, `cub unit import --dry-run`, `cub unit import --wait`, `cub unit get -o yaml`, and concrete `--where-resource` filter expressions with multiple realistic examples. Every step in the workflow has a copy-paste-ready command with placeholder patterns that are clear and consistent.

The 6-step loop is clearly sequenced with explicit validation checkpoints: preflight gates before starting, a mandatory dry-run step (step 2) with specific things to check, a review step (step 4) with concrete items to scan for, and a verify chain at the end with 5 numbered verification commands. The stop conditions section provides clear feedback loops for error recovery, and the dry-run-then-tighten pattern is an explicit iterate-until-correct loop.

The skill references many companion skills and reference files (e.g., `references/functions-catalog.md`, `references/yaml-patterns.md`, `import-unit-granularity`, `worker-bootstrap`) which is good navigation, but no bundle files are provided to verify these exist. The main content is fairly long (~180 lines of substantive content) and some sections like the detailed scoping examples or the granularity decision could potentially be split out. The references section at the end is well-organized.