Content
100%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 tight, well-structured skill body: concrete executable commands with exit-code semantics, a clear multi-step workflow with validation and feedback loops, and clean one-level-deep references that all resolve to real bundle files. No padding, no over-explanation.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean and efficient throughout — opens with a pointed thesis, never explains concepts Claude already knows, and every editorial line ('Posts without pillars are noise', 'the one they repeat beats the one that scores higher once') encodes a real decision rather than padding. | 5 / 5 |
Actionability | Copy-paste-ready commands with real flags and exit-code semantics — 'python3 scripts/format_picker.py --goal authority ...', 'python3 scripts/post_linter.py --input draft.md --has-image', exit 0/2/3 — covering the common cases fully rather than the minor-gap level of a 4. | 5 / 5 |
Workflow Clarity | Five-step sequence with explicit validation checkpoints (linter exit 0 SHIP / 2 REVISE / 3 REWRITE) and a stated stop condition ('exit 0 or a stated decision to accept a warning'); the batch repurpose operation carries ledger validation, so the destructive/batch cap does not apply. | 5 / 5 |
Progressive Disclosure | Overview body points to one-level-deep references — four references/, two assets/, three scripts/ — each clearly signaled with a description, and every linked path resolves to a real file in the bundle. | 5 / 5 |
Total | 20 / 20 Passed |