Conciseness

The body is dense and table-driven with no textbook explanation of concepts Claude already knows (e.g., it states 'ISO 42001 follows the Annex SL high-level structure shared with ISO 9001 / 27001 / 13485' as one line of working context, not a tutorial). Minor trims exist: the 38-item Keywords section and the three 'This skill is NOT...' paragraphs partially duplicate the frontmatter description, which is what keeps it below anchor 5.

Actionability

Quick Start commands are copy-paste ready and run standalone ('python scripts/aims_gap_analyzer.py # embedded sample (mid-stage AI SaaS)'), and verification of the scripts confirms each is argparse-documented with an embedded sample and a --output json flag, so the common cases are covered without any preparation. The Output Standards block specifies the exact response shape, matching the fully-executable anchor.

Workflow Clarity

Four workflows each carry a goal, timeline, and numbered steps, with confirmation checkpoints present ('For each high/critical risk, confirm >= 1 Annex A control is selected as treatment', 'Confirm auditor independence per assignment'). It stays at anchor 4 rather than 5 because no workflow includes an explicit validate-fix-retry feedback loop, though these are read-only analysis/planning tasks so the destructive/batch cap of 3 does not apply.

Progressive Disclosure

The body is a decision-framed overview; the four referenced files (references/iso42001_clauses.md, aims_controls_annex_a.md, aims_implementation_guide.md, cross_framework_mapping_ai.md) all exist and are exactly one level deep, each clearly signaled at the point of use with its purpose ('for the full clause-by-clause walkthrough with audit evidence expectations') plus a consolidated References section. Inline tables stay at overview altitude, matching the clear-overview anchor.

Specificity

The description lists multiple concrete actions with comprehensive coverage: 'Where are the gaps against Clauses 4-10 and what do we close first', 'What goes in the AI risk register and which Annex A controls treat each risk', and 'the 12-month internal audit plan that satisfies Clause 9.2'. It names the domain precisely and stays in third person, so no voice penalty applies; anchor 4 would require a coverage gap, and none is present.

Completeness

It explicitly answers both what ('AIMS specialist for compliance teams running internal audits' plus three enumerated decisions) and when ('Use when preparing for certification, scoping internal audit cycles, or onboarding AI systems into an existing ISMS (27001) / QMS (13485) program'). This matches the anchor-5 pattern of concrete what + explicit trigger phrases; the 'Use when' clause is present, so the cap of 3 does not apply.

Trigger Term Quality

Strong natural keyword coverage including synonyms: 'ISO/IEC 42001:2023', 'AI Management System (AIMS)', 'internal audits', 'AI risk register', 'Annex A controls', 'certification', 'ISMS (27001) / QMS (13485)'. A few natural phrases users would say — 'gap analysis', 'AI compliance', '42001 audit' — appear only in the body's Keywords section, not the description itself, keeping it at anchor 4 rather than 5.

Distinctiveness Conflict Risk

Clear niche (voluntary AI management-system standard, internal audit audience) reinforced by explicit negative boundaries: 'NOT an executive AI strategy skill (see chief-ai-officer-advisor). NOT EU AI Act compliance (see compliance-team-eu-ai-act)'. Minimal risk of triggering for adjacent skills; anchor 4 would require some residual overlap concern, and the disambiguation addresses the closest neighbors directly.