Conciseness

The body is lean and avoids explaining concepts Claude already knows; the only redundancy is the Examples section restating commands already shown in Instructions, a minor trim opportunity rather than padding.

Actionability

It provides concrete executable commands (roborev review --wait --type design, the heredoc+git rev-parse snippet, roborev status/init) covering the common cases, with only the placeholder substitutions (<commit>, [--panel <name>|none]) keeping it from being fully copy-paste ready.

Workflow Clarity

A clear 4-step sequence with an explicit validation gate (git rev-parse, 'do not proceed' on failure) and an error-recovery branch in step 3; the recovery is suggest-based rather than a tight fix-and-retry loop, a minor gap.

Progressive Disclosure

Well-organized into clear sections (Usage, Sandbox access, When NOT to invoke, Instructions, Examples, See also) with one-line cross-skill pointers; no bundle files exist, and the inline Panels/Sandbox content is reasonably placed for a skill of this size.

Specificity

The description 'Use only when the user explicitly invokes $roborev-design-review' names no concrete actions or domain at all — it states only when to trigger, with zero 'what does this do' content, matching the 'entirely vague; no concrete actions' anchor.

Completeness

It clearly answers 'when' (explicit invocation) but provides no 'what' — the skill's purpose (request a design review, present verdict/findings) is absent, matching the 'only when is present without what' anchor.

Trigger Term Quality

The only natural keyword is 'design review' embedded inside the command name; the surrounding phrasing ('explicitly invokes', '$roborev-design-review') is mechanical rather than the natural phrases users say, so it sits just below the midpoint.

Distinctiveness Conflict Risk

Requiring literal explicit invocation of a specific command makes accidental triggering essentially impossible — a clear niche with minimal conflict risk.