Conciseness

The ~40-line body is efficient and tool-specific (e.g. "writes the full result table to `output_csv_rel` and returns preview rows") with no concept explanations Claude already knows. Not 5 because Quick Start restates the Workflow sequence and "Suggested tools" duplicates the frontmatter metadata — minor trimmable redundancy.

Actionability

Quotes: "Use `mp_search_materials` to write a CSV candidate table", "`mp_download_structure` writes conventional standard structures under `output_dir`" — concrete tool names, parameters, and behaviors are given, but there is no example query, criteria set, or field list; "Define search criteria and requested fields" is a high-level hint without executable specifics, matching anchor 3 (concrete guidance but missing key details).

Workflow Clarity

The three steps (search with explicit criteria → prune before download → download retained candidates) are clearly sequenced, and validation checkpoints exist ("returns preview rows for quick inspection", "Use the returned `results` and `errors` rather than assuming every requested ID resolved"), so the batch-operation cap at 3 does not bind. Not 5 because there is no error-recovery feedback loop (e.g. retry or trim failed IDs) for the batch download step.

Progressive Disclosure

The skill is under 50 lines, needs no external references (no references/, scripts/, or assets/ bundles exist), and is organized into clear sections (Overview, Quick Start, Suggested tools, Workflow, Output Contract, References). Per the rubric's simple-skill guidance, well-organized sections alone merit a 5; the References section is a pointer to the tool schema, not a broken file reference.

Specificity

Quotes: "materials discovery and bulk structure selection before slab construction", "database query strategy, candidate filtering, and export readiness" — several specific activities are named, matching anchor 4. Not 3 because more than 1-2 concrete actions are listed; not 5 because downstream actions (shortlisting, structure download) are not enumerated.

Completeness

Quotes: "Use this skill for materials discovery and bulk structure selection... including database query strategy, candidate filtering, and export readiness" — the 'what' is clear and multi-part, and an explicit 'Use this skill for...' trigger clause with a 'before slab construction' condition provides the 'when'. Not 5 because the 'when' is not framed as concrete user-trigger phrases (e.g. 'when the user asks to find bulk structures').

Trigger Term Quality

Quotes: "materials discovery", "bulk structure selection", "slab construction", "candidate filtering" — good natural keyword coverage for the domain, but common variations users might say (e.g. "Materials Project", "screening", "download structures") are missing, so anchor 4 rather than 5.

Distinctiveness Conflict Risk

Quotes: "bulk structure selection before slab construction" — a clear niche with distinct domain triggers and an explicit boundary ('before slab construction'). Not 5 because "materials discovery" is broad enough to risk minor overlap with adjacent materials-science or structure-download skills.