Content
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A highly actionable, well-sequenced testing skill with executable examples and clear TDD validation loops, but it is a monolithic ~800-line document that repeats some commands and would benefit from splitting detailed reference material into separate files.
Suggestions
Move the matcher catalog, full spec-style reference, and CI/CD YAML into separate files under references/ (e.g. MATCHERS.md, SPEC-STYLES.md, CI.md) and link to them from a concise overview.
Deduplicate the Kover/gradle commands so each appears once; reference the single source from the Quick Reference section.
Tighten the Quick Reference to genuine pointers and trim repeated inline examples to reduce token cost.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The prose is lean and avoids explaining basics Claude knows, but the ~800-line body repeats commands (e.g. `koverHtmlReport` in three places) and inlines extensive reference material that could be tightened. | 2 / 3 |
Actionability | Provides fully executable, copy-paste-ready Kotlin code and concrete `./gradlew` commands for every pattern — specs, mocking, coroutines, property tests, Kover, Ktor, and CI. | 3 / 3 |
Workflow Clarity | The TDD workflow is explicitly sequenced (RED-GREEN-REFACTOR) with validation checkpoints — "Run tests - verify FAIL" and "Run tests - verify PASS" — and the How-It-Works steps include verifying expected failure before implementing. | 3 / 3 |
Progressive Disclosure | Well-organized into sections, but all detailed reference material (every spec style, full matcher catalog, CI YAML) is inline in SKILL.md with no bundle files; content that should be split into separate reference files is monolithic. | 2 / 3 |
Total | 10 / 12 Passed |