Conciseness

The Actor tables, commands, and step instructions are lean and instructional, but the frontmatter description is repeated verbatim as the body's opening paragraph, and the "Execution and export boundaries" section is one dense padded paragraph mixing many unrelated caveats (including "The repaired runner", a reference to internal history that does not exist in this skill). Mostly efficient with some tightening needed, rather than 'noticeably verbose' across several padded sections.

Actionability

The mcpc and node commands are concrete and copy-paste ready with explained placeholders ("Replace SEARCH_KEYWORDS with 1-3 simple terms", "Replace ACTOR_ID with the selected Actor"), and error handling maps each failure to a specific remedy. Minor gaps keep it from 5: --input 'JSON_INPUT' is never exemplified with a realistic value, and step numbering for "Number of results" is vague.

Workflow Clarity

Five clearly sequenced steps with a copyable progress checklist, pre-run checkpoints (fetch schema, check pricing box, "Get explicit approval", "Set a conservative result cap"), and an error-handling section with recovery actions including timeout reconciliation before retry. Minor gaps: no explicit post-run verification of output correctness, and step 3's "Based on character of use case" is ambiguous — so it sits at 'most checkpoints present' rather than the fully validated anchor at 5.

Progressive Disclosure

Section structure is clear (When to Use, Prerequisites, Workflow, Error Handling, Limitations), but the body references "$SKILL_PATH/reference/scripts/run_actor.js" and that file does not exist in the bundle (no references/, scripts/, or assets/ directories are present), and the path uses the nonstandard "reference/" spelling. ~150 lines of Actor tables are inlined rather than split into a reference file, so organization is adequate but not well-signaled or complete.

Specificity

"Select a candidate Apify Actor for the requested extraction after checking its current schema, provenance, pricing and limits" names the domain and one concrete selection/verification action, but coverage is not comprehensive — running Actors, output formats, and multi-platform coverage that the body actually provides are absent. The second sentence is a disclaimer, not a capability, so this fits the 'domain plus 1-2 concrete actions' anchor rather than the 'several specific actions' anchor above.

Completeness

The 'what' is stated clearly (select a candidate Apify Actor after checking schema, provenance, pricing, limits), but there is no 'Use when...' clause or equivalent explicit trigger guidance anywhere; 'when' is only weakly implied by "for the requested extraction". Per the rubric guideline, a missing 'Use when...' clause caps completeness at 3.

Trigger Term Quality

"Apify Actor" and "extraction" are relevant keywords, but the natural phrases users would actually say — "scrape", "web scraping", "scraper", or platform names like Instagram, TikTok, Facebook, Google Maps — are entirely missing. This matches 'some relevant keywords but missing common variations or synonyms' rather than score 2, since 'Apify' is a strong, specific (not generic) trigger term.

Distinctiveness Conflict Risk

"Apify Actor" scopes a clear niche that is unlikely to trigger for generic scraping, PDF, or document skills, with only minor overlap risk against other Apify-specific skills. The broadly applicable word "extraction" and the absence of an explicit trigger phrase keep it from the 'clear niche with distinct triggers' anchor at 5.