Conciseness

The body is mostly efficient commands and concrete steps with little padding; only minor instances could be trimmed, such as the 'guidelines, not a rigid script' boilerplate and the conceptual sandbox-access prose, matching 'Efficient; minor instances of over-explanation that could be trimmed.' Not 5 because of those small trimmable passages; not 3 because the bulk is tight and actionable.

Actionability

Provides concrete executable commands ('roborev review --branch --wait ...') and a copy-paste base-branch snippet with 'git rev-parse --verify', plus worked examples with real job ids, matching 'Mostly executable guidance; concrete code or commands with minor gaps.' Not 5 because some options are shown as template placeholders ('[--type <type>]') with conditional framing rather than a single ready command.

Workflow Clarity

A clear 4-step sequence (validate -> run -> present -> offer next steps) with an explicit validation checkpoint ('git rev-parse --verify', 'If validation fails ... Do not proceed') and error-recovery feedback loops (suggest 'roborev status'/'roborev init'/re-run, and offer '$roborev-fix'), matching 'Clear sequence with explicit validation steps; feedback loops for error recovery.' Not 4 because both validation and recovery loops are explicitly present; the destructive/batch cap does not apply since this is a read-only review.

Progressive Disclosure

Well-organized into clear sections (Usage, Instructions with numbered subsections, Examples, See also) and self-contained with no bundle files present, so navigation is easy and nothing is buried, matching 'Good structure; most content is appropriately placed; minor organization gaps.' Not 5 because there are no one-level-deep references to signal (none are needed), so it does not demonstrate the reference-splitting the top anchor rewards.

Specificity

The description names the domain and a minimal action via the invocation token '$roborev-review-branch' (review + branch), but the prose itself states no concrete capability beyond the 'Use only when' gate, matching 'Names the domain but actions are minimal or generic.' Not 3 because no additional concrete actions are listed; not 1 because the token does name the action and domain rather than being entirely vague.

Completeness

The description gives an explicit 'when' ('Use only when the user explicitly invokes $roborev-review-branch') but no explicit 'what' statement of capability; the action is only implied by the token, matching 'only when is present without what.' Not 3 because the 'what' is not clearly stated, only weakly implied by the name.

Trigger Term Quality

Natural keywords 'review' and 'branch' appear in the invocation token, but there is no coverage of common synonyms or variations a user might say (audit, inspect, check commits, PR), matching 'Some relevant keywords but missing common variations or synonyms.' Not 4 because only two relevant terms are present with no broader coverage.

Distinctiveness Conflict Risk

Triggering is gated to a single named invocation token ('$roborev-review-branch') with an explicit-invocation requirement, giving a clear niche with minimal conflict risk. Not below 5 because no other skill would match this exact explicit trigger.