Content
71%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.
A strong, highly actionable body with excellent parameter/output/error tables and worked examples, undermined by inconsistent references to its own bundled script (three different paths, one wrong) and some duplicated content between Quick Start, Workflow, and Script Location sections. Fixing the script path to its actual bundle location (scripts/jobspy_scraper.py) and de-duplicating would lift it substantially.
Suggestions
Reference the bundled script at its actual path (scripts/jobspy_scraper.py relative to the skill root) everywhere, and remove the contradictory 'skills/linkedin-scraper/scripts/...' path in the Script Location section — currently the body gives three different locations for the same file.
De-duplicate: the Quick Start run command repeats the Step 2 template and the Script Location section repeats Step 3's fallback instruction; merge these into one place to save tokens.
Standardize the interpreter invocation — the install step uses python3.12 while run examples use python — and add a brief note on where to find a LinkedIn company ID, since the company-monitoring use case currently assumes the user has one.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dominated by efficient tables and executable commands with no padding explaining known concepts, but the Quick Start command duplicates the Workflow template and the 'Script Location' section restates Step 3's instructions, so a few tokens could be trimmed. Fits anchor 4 (efficient, minor over-explanation to trim) better than 5 because of the duplicated install/run and script-location content. | 4 / 5 |
Actionability | Commands are copy-paste ready with full parameter, output-column, and error-fix tables plus three worked use cases, but the script is referenced via three inconsistent paths (tools/, skills/linkedin-job-scraper/scripts/, and prose saying skills/linkedin-scraper/scripts/) and the interpreter flips between python3.12 and python — minor gaps that keep it below fully-executable. Anchor 4 ('mostly executable; concrete commands with minor gaps') is the best fit; not 5 because of the path and interpreter inconsistencies. | 4 / 5 |
Workflow Clarity | The four-step workflow is clearly sequenced with defaults guidance, a zero-results feedback loop, and an error-recovery table; scraping is read-only so the destructive-cap doesn't apply, but validation checkpoints are implicit (error table) rather than explicit steps in the sequence. Anchor 4 ('clear sequence with most checkpoints present; minor validation gaps') fits; not 5 because there is no explicit validate-and-retry step in the workflow itself. | 4 / 5 |
Progressive Disclosure | Sections are well organized and tables are appropriately inline for a skill this size, but the only bundle file (scripts/jobspy_scraper.py) is never referenced at its actual location — the body points to tools/jobspy_scraper.py and a wrong skills/linkedin-scraper/ path, so the reference to bundled material is mis-signaled. Anchor 3 ('references present but not clearly signaled') fits; not 4 because the script reference is actively inconsistent across three paths rather than merely having minor organization gaps. | 3 / 5 |
Total | 15 / 20 Passed |