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: lean, non-obvious guidance with concrete examples, explicit checkpoints, and error-recovery loops throughout. There is nothing meaningful to trim or restructure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean and efficient with zero padding: every paragraph delivers non-obvious, environment-specific guidance ('Each repository is searched separately, so covering the whole set is slow and rarely needed') and nothing explains concepts Claude already knows. | 5 / 5 |
Actionability | Fully actionable for an instruction-only skill: quoted request patterns ('which repos mention X'), sample reports ('Three files reference `OrderStatus`'), concrete error classes ('a missing path, a read-only error, a per-command limit'), and specific fallbacks ('try `grep` before saying nothing matched') cover the common cases. | 5 / 5 |
Workflow Clarity | Each rule is unambiguous and includes checkpoints and feedback loops: user confirmation before batch repository scanning, a `grep` fallback before reporting no-match, and fix-and-retry on command failure ('fix it and retry: the error says how'). | 5 / 5 |
Progressive Disclosure | Under 50 lines with well-organized bold-headed sections and no bundle files; command mechanics are cleanly delegated in one clearly signaled step ('The `exploreRepo` description covers the mounts, the commands and the `search` locator'). | 5 / 5 |
Total | 20 / 20 Passed |