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.
An exemplary routing-style skill: concise, entirely composed of non-obvious domain rules, with concrete right/wrong examples and a clean one-level-deep reference structure that matches the actual bundle. The only gap is a missing post-edit verification checkpoint, which keeps workflow clarity at 4 rather than 5.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and every token is domain knowledge Claude cannot infer (Block Editor bare-variable exception, responseBodySchema-only response fields, empty-array vs empty key/value entries). It assumes competence and includes no padding or generic explanation of Novu, Liquid, or notifications. | 5 / 5 |
Actionability | Guidance is fully concrete and copy-paste ready: exact variable syntax ('{{ subscriber.firstName }}'), explicit wrong/right examples ('"{{ payload.actionUrl }}"' is wrong; '"payload.actionUrl"' is right), a worked steps.fetch-user.name example, and exact JSON for merge/replace/null condition semantics, covering the common cases per step type. | 5 / 5 |
Workflow Clarity | The routing sequence is clear and coherent ('Open the matching reference before writing controls', when-to-use examples, add-vs-replace decision rules), but there is no validation checkpoint after writing step content (e.g., confirming the workflow validates or renders correctly) — a minor checkpoint gap. | 4 / 5 |
Progressive Disclosure | The SKILL.md is a clear overview that routes each of the ten step types to a real one-level-deep reference file via a well-signaled table, keeps common pitfalls inline, and cross-links sibling skills — easy navigation with content appropriately split. | 5 / 5 |
Total | 19 / 20 Passed |