Content
100%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 instruction-only skill body: the red-green-refactor loop is sequenced with explicit validation checkpoints and feedback at every step, the test-quality rules are concrete and skill-specific rather than generic testing tutorials, and the whole thing is tight at 44 lines with no bundle sprawl.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean throughout — no padding, no explaining what testing is; every sentence adds a skill-specific judgment rule ("a test that mocks the thing under test asserts nothing"). Matches the 5 anchor: every token earns its place. | 5 / 5 |
Actionability | Concrete, directly followable instruction at every step — "Confirm it fails for the right reason", "Write the least code that makes the test pass", "Mock at the system boundary (network, clock, filesystem)", with `expect(x).toBe(x)` as a concrete anti-pattern. Per the rubric's code_vs_instruction note, the absence of code in an instruction-only skill is not penalized. | 5 / 5 |
Workflow Clarity | The Red → Green → Refactor → Repeat loop is clearly sequenced with explicit validation checkpoints at every step ("Run it. Confirm it fails for the right reason", "Run the test. Green.", "Re-run after every edit") plus error-recovery guidance (fix test setup first if it errors). Matches the 5 anchor. | 5 / 5 |
Progressive Disclosure | A 44-line self-contained skill with no bundle files and clear sections (The loop / What makes a test worth writing / When to reach for it / Output contract). Per the rubric's guideline for skills under 50 lines with no need for external references, well-organized sections alone warrant a 5. | 5 / 5 |
Total | 20 / 20 Passed |