Content
92%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.
The content is highly actionable with executable commands, clear validation-gated workflows for destructive operations, and well-structured progressive disclosure via a real reference bundle. The only weakness is mild redundancy in the quick-decision section that could be tightened.
Suggestions
Consolidate the repeated node-get URL-to-space_id resolution steps into a single canonical snippet referenced from each bullet, to remove redundancy in the '快速决策' and delete-space sections.
Consider moving the lengthy delete-space resolution procedure into its reference file and keeping only the safety constraint ('必须先解析出 space_id 并由用户选定') inline.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and assumes Claude's competence with no concept-explaining fluff, but the node-resolution steps and safety restatements are repeated across several '快速决策' bullets and the delete-space section, leaving minor trimmable redundancy. | 4 / 5 |
Actionability | Fully executable, copy-paste-ready commands throughout—e.g. 'lark-cli wiki +node-get --node-token \'<wiki_url>\' --as user --format json' and 'lark-cli wiki +delete-space --space-id <ID> --yes'—covering the common cases with concrete flags and JSON params. | 5 / 5 |
Workflow Clarity | Destructive/batch operations are sequenced with explicit validation checkpoints: delete-space requires resolve → list candidates to user → user selects → '--yes', with 反例 (anti-examples) and error-handling rules distinguishing rate_limit retries from invalid_parameters fixes. | 5 / 5 |
Progressive Disclosure | SKILL.md is an overview with a shortcut table linking to 13 verified one-level-deep reference files under references/, keeping detail out of the body while providing easy navigation; cross-skill references are contextual and single-level. | 5 / 5 |
Total | 19 / 20 Passed |