Content
96%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 tightly written, highly actionable skill body with an exemplary validated workflow. Its one real weakness is progressive disclosure: the body references six bundle files that do not exist in the package, breaking the navigation it so clearly signals.
Suggestions
Ship the referenced bundle so the signaled links resolve: scripts/post_performance_analyzer.py, scripts/pattern_miner.py, scripts/experiment_planner.py, references/linkedin_metrics_canon.md, references/evidence_thresholds.md, assets/example_post_export.csv, and assets/measurement_log_template.md — none are present in the package.
If a file is intentionally not bundled, mark it as such inline so Claude does not attempt to follow a dead link.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean and dense; assumes Claude's competence and skips generic explanations (no 'what is a median'), with every explanatory clause tied to domain-specific statistical reasoning that earns its place. | 5 / 5 |
Actionability | Copy-paste-ready commands with full flags and example values for all three scripts (e.g. '--input posts.csv --csv --output human', '--cv 0.45 --effect 0.30 --posts-per-week 2'), covering the common cases concretely. | 5 / 5 |
Workflow Clarity | A clear four-step sequence (export → describe → test → experiment) with explicit exit-code checkpoints (0/2/3) at each stage serving as validation feedback loops, including refusal gates for the under-10-posts floor. | 5 / 5 |
Progressive Disclosure | Well-structured overview with a scripts table, clearly signaled one-level-deep references, and a 'Distinct from' section — but the six referenced bundle files (scripts/, references/, assets/) are not present on disk, so the signaled navigation points to missing materials. | 4 / 5 |
Total | 19 / 20 Passed |