Content
100%Weight 40%Scale 1-3Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A lean, well-structured reference-card skill that fully executes its single task: it tells Claude exactly what to display, what to avoid (no persistence), and lists every trigger/mode concretely. No redundancy or padding.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is terse and table-driven, explains nothing Claude already knows, and every line earns its place; even the config examples are minimal and load-bearing. | 3 / 3 |
Actionability | It gives concrete, copy-paste-ready guidance — exact trigger strings, exact mode behaviors, an `export CAVEMAN_DEFAULT_MODE=ultra` command, and a config JSON snippet — plus an unambiguous 'Display this reference card... do NOT change mode, write flag files, or persist anything' instruction. | 3 / 3 |
Workflow Clarity | As a simple single-action skill ('display this reference card when invoked, one-shot'), the action is unambiguous and bounded with explicit do-nots; per the simple-skills note this clears the top anchor. | 3 / 3 |
Progressive Disclosure | Under 50 lines with no need for external references and no bundle files present; content is split into clearly labeled sections (Modes, Skills, Deactivate, Configure, More), which satisfies the well-organized-sections allowance for simple skills. | 3 / 3 |
Total | 12 / 12 Passed |