Conciseness

The ~85-line body is lean and assumes competence: there is no explanation of what LinkedIn is or how posts work; every line is a command, an exit-code contract, a rule, or a scope boundary, and even the stylistic one-liners ("Posts without pillars are noise") carry routing signal. Not 4: no section could be trimmed without losing operative content.

Actionability

Three fully executable commands with complete flags (e.g. "python3 scripts/post_linter.py --input draft.md --has-image --output human"), explicit exit-code semantics (0 SHIP / 2 REVISE / 3 REWRITE) with named consequences, a concrete fold budget ("the first ~140 characters"), and concrete drafting guidance ("the number, the mistake, the sentence someone said"). Not 4: the commands are copy-paste ready and the common cases are covered.

Workflow Clarity

A five-step sequence with an explicit validation feedback loop: lint before shipping with exit-code outcomes, "Two passes is normal; the stop condition is exit 0 or a stated decision to accept a warning", plus tie-break and refusal rules in the format picker and a brief-check gate in step 1. This matches the anchor with validate → fix → retry and a stated stop condition.

Progressive Disclosure

A clear overview that splits detail into one-level-deep, well-signaled resources: three scripts listed in a table with roles, four references each annotated with its scope ("the first 140 characters, and the openers to delete on sight"), and two assets — navigation is easy and nothing is buried. Note: no bundle files were present in the review directory to verify the paths against, but the structure as written is the reference ideal. Not 4: content is appropriately split with clear signals throughout, with no inlined bulk that belongs elsewhere.

Specificity

Lists multiple specific concrete actions with comprehensive coverage: "write, edit, or lint a LinkedIn post", "Lints posts 0-100 on mechanics, hook, integrity, and accessibility", "picks the format the material actually supports", and "splits long sources into standalone units with a reuse ledger" — every capability of the skill is named with quantified detail. Not 4: there are no gaps in coverage; the actions span drafting, linting, format choice, and repurposing.

Completeness

Explicitly answers both questions: "Use when someone wants to write, edit, or lint a LinkedIn post... Triggers on..." (when, with concrete trigger phrases) and "Lints posts 0-100...; picks the format...; splits long sources..." (what, in operational detail). Not 4: neither half is missing, implicit, or underspecified.

Trigger Term Quality

Comprehensive natural trigger phrases quoted verbatim — "write a LinkedIn post", "is this hook any good", "review my post", "turn this into LinkedIn posts", "carousel", "what format should this be" — plus synonyms for every content type (story, how-to, opinion piece, carousel script, video script, poll, article, talk, transcript). Not 4: these are exactly the phrases a user would naturally say, not just keyword labels.

Distinctiveness Conflict Risk

A clear niche (LinkedIn posts specifically) with distinct triggers; nearly every trigger phrase contains "LinkedIn" or an unambiguously LinkedIn-native artifact ("carousel", "post"), so the wrong-skill trigger risk is minimal. Not 4: the closely related general-copywriting skills are disambiguated by the platform scoping, not merely "mostly" distinct.