Conciseness

The body is a heading plus one sentence — "This skill can only be invoked via /skill:disable-model-invocation." — with zero padding or explanation of concepts Claude already knows. Every token earns its place, matching the "Lean and efficient" anchor.

Actionability

The single sentence is concrete (it names the exact slash command), but the skill defines no capability, task, or steps for Claude to execute, matching "Some concrete guidance but incomplete". It is not a 4 because there is no executable instruction of any kind, and not a 2 because the one fact given is exact rather than a high-level hint.

Workflow Clarity

This is a simple single-purpose skill under 50 lines whose single action (manual invocation) is unambiguous, so the simple-skill exception applies; however, no actual task for Claude is defined, leaving it between "clear sequence with most checkpoints" and the simple-skill-5 case. There are no destructive or batch operations, so no validation cap applies.

Progressive Disclosure

The skill is under 50 lines with no need for external references (no references/, scripts/, or assets/ exist and none are cited), and the body is a well-organized heading plus statement. Per the rubric guideline, this earns the "simple skill" 5 for progressive disclosure.

Specificity

"A skill that cannot be invoked by the model" names no domain and no concrete actions — it describes invocation mechanics rather than any capability, matching the anchor "Entirely vague; no concrete actions; pure abstract language". It cannot be a 2 because even "Processes PDF files" names a domain, which this lacks.

Completeness

The "what" is only a negative abstraction (what the skill does not do) and the "when" is entirely absent, matching "Missing both what and when, or both are extremely vague". The missing "Use when..." clause also caps completeness at 3, but the text fails even the vague-what threshold.

Trigger Term Quality

The description contains no natural keywords a user would say; "cannot be invoked by the model" is meta/technical language about the harness, not trigger phrasing. It sits at the anchor "No natural keywords; only technical jargon or entirely generic language".

Distinctiveness Conflict Risk

The phrasing is generic and describes no niche, but the explicit restriction "cannot be invoked by the model" does distinguish its invocation mode from ordinary skills, placing it between "Entirely generic" (1) and "Very broad; high overlap risk" (2). It is not a 3 because nothing in the text identifies a distinct capability or trigger.