Content
75%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 excellent executable commands and real one-level-deep references. Its weaknesses are redundancy (duplicated plugin-lookup workflow, repeated commands, restated description) and an inline Anti-Patterns section that inflates token cost without adding proportional value.
Suggestions
Deduplicate the plugin documentation lookup guidance: keep it in one place (e.g., the Plugin Lookup Workflow section) and reduce Assistant Workflow step 3 to a one-line pointer, removing the repeated MCP/WebSearch details.
Move the Anti-Patterns section into a reference file (e.g., references/anti_patterns.md) and keep 3-4 of the highest-risk items inline as a brief list, cutting roughly 50-60 lines from the main file.
Merge the Quick Start commands into the Command Reference section and delete the verbatim restatement of the frontmatter description at the top of the body.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient (terse capability lists, command-focused sections), but includes noticeable redundancy: the plugin documentation lookup workflow is stated twice in near-identical detail (the 'Plugin Lookup Workflow' section and again in Assistant Workflow step 3), command invocations are repeated between Quick Start and the Command Reference, and line 8 restates the frontmatter description verbatim. The ~70-line Anti-Patterns section could be tightened. | 3 / 5 |
Actionability | Fully executable guidance throughout: copy-paste-ready bash commands covering all flag variants (--syntax-only, --security-only, --strict, etc.), a script architecture map, an executable Declarative skeleton, concrete MCP tool names with example queries, and a working validate/fix/re-run example. Specific examples cover the common cases. | 5 / 5 |
Workflow Clarity | The Assistant Workflow is a clear 5-step sequence with explicit reporting requirements (line numbers, severity, corrected snippets), and validation feedback loops exist (anti-pattern guidance to re-run validation after fixing credentials, and to stage-test rather than trust linter results). However, the main workflow itself lacks an explicit 'apply fixes then re-validate' step — the loop lives only in the anti-patterns — leaving a minor checkpoint gap. | 4 / 5 |
Progressive Disclosure | Good structure: six one-level-deep reference files that all exist and are clearly signaled with per-file descriptions, an inline script-architecture tree, and a References index. The gap is that the long inline Anti-Patterns section and the duplicated plugin-lookup instructions are bulk content that belongs in a reference file, keeping the main file longer than an overview needs to be. | 4 / 5 |
Total | 16 / 20 Passed |