Conciseness

The ~40-line body is lean throughout: the factor table replaces prose, the trusted-source warning is one line, and no space is spent explaining concepts Claude already knows. Every token earns its place, matching the 5 anchor.

Actionability

For an instruction-only skill the guidance is fully executable: exact source fields (e.g. "nationality_or_jurisdiction", "beneficial_owners", "documents_received"), a copy-paste-ready output JSON block, enumerated output values ("low | medium | high", "pass | fail | n/a"), and hard rules ("Cite the rule — no outcome without a rule reference"). This matches the 5 anchor's specificity without needing code.

Workflow Clarity

A clear 4-step sequence (rate → check documents → rule outcomes → disposition) ending in an explicit gate ("clear only if rating is low/medium, all required docs received, and no escalation rule fired"). It sits at 4 rather than 5 because error-handling checkpoints are absent — no guidance for conflicting or unavailable inputs such as a screening result that contradicts pep_declared, a missing rules grid, or how to treat expired vs. missing documents differently in the disposition.

Progressive Disclosure

The skill is under 50 lines, needs no external references (its variable inputs — the rules grid and screening results — are runtime data, not skill documentation), and is organized into clearly headed sections. Per the simple-skill guideline, this scores 5.

Specificity

"assign a risk rating, list every rule outcome with the rule cited, and flag what's missing or escalation-worthy" names the domain and multiple concrete, distinct actions covering the skill's full output surface. It matches the 5 anchor (multiple specific concrete actions, comprehensive coverage) rather than 4, which is reserved for actions with coverage gaps.

Completeness

Both "what" (three concrete outputs) and "when" ("Use after kyc-doc-parse") are explicit, but the "when" is limited to the upstream pipeline trigger and omits user-intent phrases like "use when onboarding a client". This is the 4 anchor ("when" could be more explicit or specific); it is above 3 because an explicit Use clause exists.

Trigger Term Quality

Natural terms like "KYC/AML", "onboarding record", "risk rating", and "escalation" are present, but "Use after kyc-doc-parse" is a pipeline pointer rather than a phrase a user would say, and common synonyms ("know your customer", "screening", "due diligence", "sanctions") are absent. This fits the 4 anchor (good keyword coverage, a few natural terms missing), not 5 (comprehensive coverage including synonyms).

Distinctiveness Conflict Risk

A clear niche — post-parse rules-grid scoring for KYC onboarding — with an explicit boundary ("this skill decides nothing, it scores and routes") separating it from its upstream parser. Minimal conflict risk with other skills, matching the 5 anchor.