Content
93%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
Excellent, tight content for a simple single-purpose skill: executable commands, a complete CLI reference, cost transparency, and honest caveats, with the implementation properly externalized to the bundled script. The only gap is the absence of verification guidance for batch multi-profile runs (empty results, timeouts, token failures).
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~44-line body is lean and assumes Claude's competence: three executable commands, a CLI flag table, a one-line cost note, and two caveats — no padding and no explanation of concepts Claude already knows. Nothing could be cut without losing information. | 5 / 5 |
Actionability | Fully executable, copy-paste-ready commands covering the common cases (single profile, multi-profile with keyword/date filtering, summary output), plus a complete flag reference with defaults and a cost estimate. The referenced script exists in the bundle and its flags match the documented CLI. | 5 / 5 |
Workflow Clarity | The single action (run the script) is unambiguous with concrete commands, but the skill supports batch runs over multiple profiles with no verification guidance — nothing on handling empty results, timeouts, or partial profile failures. It is not a 3 because the one-step workflow is clear and the operation is read-only; it is not a 5 because those minor validation gaps exist. | 4 / 5 |
Progressive Disclosure | Under 50 lines with no need for external references; well-organized sections (Quick Start, CLI Reference, Cost, Notes) and the implementation correctly split into scripts/scrape_linkedin_posts.py, which is referenced by real paths. Per the rubric's simple-skill guidance this earns a 5. | 5 / 5 |
Total | 19 / 20 Passed |