Content
68%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 highly actionable, code-first reference with executable examples for every major Jest pattern, but it reads as a pattern catalog rather than a guided workflow. It also spends tokens on material Claude already knows (basic matchers, generic best practices) instead of project-specific conventions.
Suggestions
Replace the implicit sequence with an explicit workflow: generate the test file per the naming map → run it (bun test <file>) → verify it passes → iterate, and wire the Quality Checklist in as the validation step at the end of that loop.
Cut or move the "Common Matchers" section and the generic "Best Practices" list — Jest matchers and practices like "keep tests independent" are knowledge Claude already has; keep only project-specific conventions.
Clarify the runner story: the skill is named "jest-generator" and the examples use jest.*, but the preferred command is "bun test" (Bun's own jest-compatible runner, not Jest). State which runner is expected and when to fall back to npm test, since running generated jest.mock-based tests under the wrong runner can silently differ.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The presentation is terse and code-first with no prose padding, but the "Common Matchers" section (toBe, toEqual, toBeNull, toContain...) and the generic "Best Practices" list ("Keep tests independent", "Use descriptive test names") are Jest knowledge Claude already has. Mostly efficient with some unnecessary sections, matching anchor 3 rather than 2 since there are no padded explanations. | 3 / 5 |
Actionability | Everything is executable: concrete naming mappings (src/utils/validator.ts → src/utils/validator.test.ts), runnable commands (bun test src/utils/validator.test.ts, npm test -- --coverage), and copy-paste-ready mocking, spy, timer, async, and React Testing Library patterns. This matches the fully-executable top anchor. | 5 / 5 |
Workflow Clarity | The pieces (naming → running → structure → checklist) imply a sequence but never state it: there is no explicit generate-tests → run → verify-they-pass → iterate loop, and the Quality Checklist sits at the end as a static list rather than an integrated validation checkpoint. Sequence present with implicit checkpoints matches anchor 3, not 4. | 3 / 5 |
Progressive Disclosure | A single-file skill (no references/, scripts/, or assets/ exist) with clear, well-ordered section headers, so navigation is easy. At ~190 lines it exceeds the under-50-line simple-skill case, and the matcher/pattern reference material could arguably be split into a separate file, keeping it at anchor 4 rather than 5. | 4 / 5 |
Total | 15 / 20 Passed |