Content
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is well-structured and appropriately concise for a simple read-only skill, with a clean single one-level reference, but it lacks a complete executable example call and an explicit numbered workflow, and restates the auto-fill rule redundantly.
Suggestions
Provide one complete, copy-paste-ready example call, e.g. `lark-cli attendance user_tasks query --data '{"user_ids":[]}' --params employee_type=employee_no`, to raise actionability.
Collapse the three restatements of the employee_type/user_ids auto-fill rule into the table plus a single reminder to remove redundancy.
Render the operating sequence as a short numbered list (Read lark-shared → run schema → build --data/--params → call) with an explicit verify step on the returned result.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is generally lean and assumes Claude's competence (no explanation of what Feishu/Lark is), but the auto-fill rule (employee_type="employee_no", user_ids=[]) is restated three times — the parameter table, the 填充示例 block, and the 注意 note — which could be tightened, matching 'Mostly efficient but includes some unnecessary explanation or could be tightened' rather than the level-3 'every token earns its place'. | 2 / 3 |
Actionability | It gives concrete CLI templates ("lark-cli schema attendance.<resource>.<method>", "lark-cli attendance <resource> <method> [flags]") and a JSON --data example, but provides no complete filled-in example call with the actual resource/method (e.g. a full `lark-cli attendance user_tasks query --data '{...}'`), so guidance is concrete yet incomplete — anchor 'Some concrete guidance but incomplete; missing key details' rather than the copy-paste-ready level 3. | 2 / 3 |
Workflow Clarity | Ordering cues exist ("MUST 先用 Read 工具读取", "必须先运行 schema"), but the steps are not presented as an explicit numbered sequence and checkpoints are implicit; this fits 'Steps listed but validation gaps; sequence present but checkpoints missing or implicit' rather than the level-3 'Clear sequence with explicit validation steps'. | 2 / 3 |
Progressive Disclosure | The skill is under 50 lines with well-organized sections (默认参数自动填充规则, API Resources, user_tasks, 权限表) and a single clearly-signaled one-level reference ("../lark-shared/SKILL.md") and no bundle files, matching the scoring note that simple skills can score 3 with well-organized sections. | 3 / 3 |
Total | 9 / 12 Passed |