Content
85%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.
A highly actionable, well-structured skill body with strong validation feedback loops and clean progressive disclosure. Its main weakness is redundancy: several hard rules are restated verbatim across multiple sections.
Suggestions
Consolidate the repeated XML/element hard rules (chart-vs-shape, <img> not <image>, table construction) into a single authoritative section and cross-reference it from 权威经验 and 核心规则 instead of restating them.
Move the dense per-shortcut parameter detail already living in references/cli/*.md out of the Shortcuts/Quick Reference tables to reduce repeated inline expansion.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Tone is lean and assumes Claude's competence (no basic-concept explanation), but the same hard constraints (chart vs shape, <img> not <image>, table rules) are repeated across 权威经验, 核心规则, and the Quick Reference table, adding tokens that could be consolidated. | 3 / 5 |
Actionability | Provides concrete executable commands (lark-cli slides +create/+add-slide/+delete-slide, auth login, wiki +node-get), ready XML snippets (gradient rgba format, <content> usage), and a need-to-command table covering common cases. | 5 / 5 |
Workflow Clarity | The 生成流程 (Steps 1–4) is clearly sequenced with explicit validation checkpoints (xml_lint.py error_count==0, +xml-get readback, validation-xml.md) and feedback loops for error recovery on destructive/XML operations. | 5 / 5 |
Progressive Disclosure | SKILL.md is a well-signaled overview pointing to a rich one-level-deep bundle (references/cli/*, references/workflow/*, references/xml/*, scripts/*) via the Quick Reference table, 执行前必做 list, and Shortcuts table; referenced paths correspond to real bundle files. | 5 / 5 |
Total | 18 / 20 Passed |