Conciseness

The body is dense, checklist-style guidance ("Replace broad 'ask the user' language with explicit choices"; "no `<...>`, 'TBD', 'TODO'") with no padding explaining concepts Claude already knows. Not a 5 because completion criteria, fallback behavior, and response format each appear twice — once in the section templates and again in the audit rubric — which could be tightened into a single cross-referenced list.

Actionability

Every required section ships with concrete Bad/Good example pairs ("Bad: 'You are a helpful AI assistant.' / Good: 'You are Acme's Tier-1 Email Triager...'") plus a copy-paste-ready completion-criteria template and an 11-item checkable audit list. Per the rubric's scoring note, an instruction-only skill without code is not penalized when the guidance is this actionable, matching the top anchor.

Workflow Clarity

The sequence is explicit — numbered run contract (1–6), nine named sections, then a mandatory audit rubric with the feedback loop "If any item fails, revise the system prompt before writing it" — and "How to use this skill" adds a four-step path for the novel-outcome case. This is prompt authoring, not a destructive/batch operation, so no cap applies; the checklist-plus-retry structure matches the 5 anchor.

Progressive Disclosure

No bundle files exist, and the single-file body is well-sectioned with clear headers while domain detail is correctly externalized to the archetype skills. Below 5 because the "Cross-references" section names sibling skills (coding-agent, research-agent, ...) as plain text rather than signaling them as navigable references, and the file exceeds the under-50-line simple-skill exception.

Specificity

Enumerates several concrete components it provides — "required run contract, completion criteria, fallback paths, response format, and anti-patterns" — in third person ("It defines"), rather than vague filler. Not a 5 because the coverage hangs off one umbrella verb ("defines") rather than a list of distinct concrete actions.

Completeness

Explicitly answers both questions: "what" ("Universal quality bar and final audit rubric for any agent system prompt... defines the required run contract, completion criteria, fallback paths, response format, and anti-patterns") and "when" with concrete triggers ("Activate this whenever you are unsure which archetype skill applies, or as a final review pass before writing the system prompt"). This matches the top anchor; the explicit 'Use when'-style clause rules out the 4 anchor's weaker 'when'.

Trigger Term Quality

Natural trigger phrases like "agent system prompt", "final review pass", and "writing the system prompt" are present and phrased the way a user would say them. Missing a few common synonyms (e.g. "prompt engineering", "agent instructions"), so it falls just short of the comprehensive 5 anchor.

Distinctiveness Conflict Risk

Clear niche — the compile-check/audit layer — explicitly positioned relative to the archetype skills ("whenever you are unsure which archetype skill applies"), which reduces conflict risk. The scope "any agent system prompt" is still broad with minor overlap against those related skills, keeping it below the minimal-conflict 5 anchor.