Conciseness

Lean and efficient throughout: the intro's three-reader model (scanner/evaluator/decider with timings) is domain expertise that justifies the workflow's ordering, not padding, and the Rules section is one tight line each. Not level 4 because nothing explains concepts Claude already knows and no section could be cut without losing guidance.

Actionability

Three copy-paste-ready bash commands with flags and documented exit codes ('--input profile.json --output human', 'Exit 0 STRONG / 2 INCOMPLETE / 3 WEAK'), plus an explicit stop condition. Not level 5 because 'Collect the five parts from the user' never names the five parts (hook, audience, proof, approach, CTA) in the body, and the input shapes of profile.json/about.json are deferred entirely to asset files.

Workflow Clarity

Four explicitly ordered steps with validation via exit codes and refusal conditions, and a real feedback loop: 'Iterate here. Two or three passes is normal; the stop condition is exit 0, or the user saying they would say it out loud to a peer.' Not level 4 because checkpoints and error-recovery loops are explicit for every scripted step.

Progressive Disclosure

The body is an overview; every referenced path (3 scripts, 2 references, 2 assets) exists in the bundle, is one level deep, and is annotated with its role ('what each section is for, and the three readers', 'input shape for the auditor'). Not level 4 because detail is appropriately split out and navigation is via a labeled table and bulleted list with no buried references.

Specificity

Multiple concrete actions with comprehensive coverage: 'audited or rewritten — headline, About section, experience bullets, Featured, banner, recommendations', plus 'Scores the headline on five dimensions, audits the whole profile 0-100 and ranks fixes by points per hour, and assembles an About section'. Not level 4 because coverage of the domain's actions is complete rather than having minor gaps.

Completeness

Explicitly answers both: what ('Scores the headline on five dimensions, audits the whole profile 0-100 and ranks fixes by points per hour, and assembles an About section that survives the "…see more" fold') and when ('Use when someone wants their LinkedIn profile audited or rewritten … or says …' with concrete trigger phrases). Not level 4 because neither what nor when could be more explicit.

Trigger Term Quality

Quotes natural user utterances verbatim — 'fix my headline', 'my profile gets views but nothing happens', 'optimize my LinkedIn profile', 'what should my About section say' — alongside synonyms (audited, rewritten, optimize, fix). Not level 4 because these are exact phrases a user would naturally say, not just generic keywords.

Distinctiveness Conflict Risk

Clear niche (LinkedIn profile audit/rewrite) with distinct triggers unlikely to fire for posting-strategy or general marketing skills. Not level 4 because the quoted trigger phrases ('fix my headline', 'what should my About section say') map unambiguously to this skill with minimal overlap risk.