Content
76%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 concise, well-organized body with executable per-platform code examples and useful best-practice/anti-pattern lists. Main weaknesses are the absence of an explicit multi-step testing workflow with validation checkpoints, and missing run/setup details that would make the examples fully actionable end to end.
Suggestions
Add a short ordered workflow (e.g., 1. write unit tests for business logic, 2. add integration/E2E for critical flows, 3. run suite, 4. triage flaky tests and rerun) to give workflow_clarity explicit checkpoints and a feedback loop.
Include the commands to actually run each framework's tests (e.g., `npx jest`, `npx detox test`, `xcodebuild test`, `./gradlew connectedAndroidTest`) so the examples are executable end to end.
Mention how test IDs are wired up (Detox `testID`, iOS accessibility identifiers, Android `viewTag`) since all three examples rely on element IDs but none show how to set them.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean with zero padding: a one-line purpose, a compact pyramid table, code examples, and terse best-practices/avoid lists. Every section adds information Claude does not already have, matching the 'every token earns its place' anchor. | 5 / 5 |
Actionability | The Jest, Detox, XCTest, and Espresso examples are concrete, syntactically complete, and copy-paste ready. Not a 5 because key execution details are missing — no test-runner commands, Detox configuration, or guidance on wiring test IDs (`by.id`, accessibility identifiers). | 4 / 5 |
Workflow Clarity | The pyramid table and 'Test business logic first' imply an ordering, but there is no explicit multi-step process for testing a feature, and no validation or feedback checkpoints (e.g., run suite -> triage flaky tests -> rerun) — matching the 'sequence present but checkpoints missing' anchor. | 3 / 5 |
Progressive Disclosure | Well-organized sections with clear headers and appropriately inline examples for a single-file skill. Not a 5 because the ~83-line body includes full per-platform example suites that could be split into one-level-deep reference files (e.g., ios-testing.md, android-testing.md) to keep SKILL.md a leaner overview. | 4 / 5 |
Total | 16 / 20 Passed |