Content
82%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 well-structured skill overview: high-value gotchas and an executable quick-reference stay inline while bulk details live in four clearly-navigated reference files. Main improvement room is trimming transitional padding and adding an explicit verify-and-fix loop in the final workflow step.
Suggestions
Cut transitional filler like 'These are the things that silently break sites. Know them before you start.' and let the gotcha list speak for itself.
Strengthen the 'Run and verify' step with what to check (e.g., page renders, admin UI loads, seed applied) and what to do on failure.
Trim the inline API cheat sheet to the few most-used calls and leave the full import surface to references/querying-and-rendering.md to reduce duplication.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is efficient and teaches only EmDash-specific facts Claude cannot know (image field objects, entry.id vs entry.data.id, cache hints), but has minor padding such as 'These are the things that silently break sites. Know them before you start.' | 4 / 5 |
Actionability | Guidance is fully executable: copy-paste-ready import/query snippets in the cheat sheet, correct `<Image image={...} />` usage, a concrete `emdash({...})` registration example, and the `pnpm dev` command. | 5 / 5 |
Workflow Clarity | The six-step workflow (configure → schema → pages → features → seed → run and verify) is clearly sequenced with a verification step, but lacks an explicit error-recovery feedback loop (what to check and fix if the site fails to render). | 4 / 5 |
Progressive Disclosure | Four real, substantive, one-level-deep reference files are clearly signaled both in the workflow steps and a summary table; the sizable inline API cheat sheet partially duplicates reference content, a minor organization gap. | 4 / 5 |
Total | 17 / 20 Passed |