Conciseness

The body is lean with zero padding and no explanation of concepts Claude already knows; the only flaw is redundancy — the opening line 'You help customers with invoices, payments, refunds, and plan changes' restates the '## Skills' bullet list verbatim. That minor trimmable redundancy places it at anchor 4 ('efficient; minor instances that could be trimmed') rather than 5's 'every token earns its place'.

Actionability

Concrete guidance is limited to two tool names with one-line descriptions ('get_invoice: Retrieve customer invoices', 'process_refund: Issue refunds'); the four skill entries ('Payment processing', 'Refund handling', 'Plan changes and upgrades') are bare labels with no steps, criteria, or examples — and two of them have no corresponding tool. This matches anchor 2 ('minimal concrete guidance; high-level hints but missing the specific steps'), not 3, because no skill area has a complete or near-complete procedure.

Workflow Clarity

There is no sequence at all: no steps for handling a refund, payment, or plan change, and no validation before irreversible financial actions like process_refund, matching anchor 1 ('steps missing or incoherent; no sequence; no validation for risky operations'). It is not 2 because even a rough sequence is absent, and the simple-skill exception does not apply since the skill declares four distinct multi-step task areas.

Progressive Disclosure

The body is under 50 lines with no external references needed (no references/, scripts/, or assets/ files exist in the bundle), and it is organized into clear '## Skills' and '## Tools' sections. Per the rubric's guideline for short, self-contained skills, well-organized sections alone warrant a 5.

Specificity

The description names the domain and lists several concrete capability areas — 'invoices, payments, refunds, and plan changes' — comparable to the anchor 'lists several specific actions; minor gaps'. It falls short of a 5 because the capabilities are stated as noun phrases rather than concrete operations (e.g., 'issue refunds against a verified invoice'), and it is above 3 because it covers four distinct capability areas rather than only 1-2.

Completeness

It explicitly answers both what ('Billing specialist for invoices, payments, refunds, and plan changes') and when ('Use when customers ask about charges, billing inquiries, or subscription management') with concrete trigger phrases, matching the anchor-5 example structure. It is not a 4 because the 'when' clause is already explicit and specific, not merely present with room to improve.

Trigger Term Quality

Trigger phrases 'charges, billing inquiries, or subscription management' are natural terms a customer would actually say, alongside 'invoices, payments, refunds'. It does not reach 5 because common variations are missing (e.g., 'overcharged', 'duplicate charge', 'cancel my plan', 'receipt'), but it clearly exceeds 3's 'missing common variations or synonyms' since charges/billing/subscription synonyms are present.

Distinctiveness Conflict Risk

Billing is a clear niche with distinct triggers, and 'typically reached via handoff from the support agent' explicitly delineates routing from adjacent skills, giving minimal conflict risk. It does not exhibit the overlap-with-similar-skills weakness of anchor 4.