Content
82%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 tight, highly actionable CLI skill: copy-paste commands throughout, a sensible step sequence with a CI exit-code checkpoint, and a well-organized single-file layout with a clearly signaled external reference. Weaknesses are minor: a few lines re-explain concepts Claude already knows and the workflow lacks an error-recovery loop after a failed gate.
Suggestions
Trim Claude-known background: drop or compress 'Lizard's CCN counts decision points (if/else/case/&&/||/?:)' and the NLOC/CCN abbreviation glosses — keep only the Lizard-specific mapping (which flags set which threshold).
Add a short feedback loop to Step 5: after a non-zero exit, show how to extract the offending functions from lizard.csv (e.g., the grep/jq pipeline teased in Step 4) and re-run after fixes.
Consider moving the Anti-patterns table's remediation detail (e.g., the diff-only `git diff --name-only` workflow) into a short referenced section or example so the main flow stays a lean overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean — commands, a flag table, and an anti-patterns table with almost no filler. Minor over-explanation of things Claude already knows ("CCN counts decision points (if/else/case/&&/||/?:)" and the NLOC/CCN abbreviation glosses) keeps it below the 'every token earns its place' anchor 5, matching anchor 4's 'minor instances of over-explanation that could be trimmed'. | 4 / 5 |
Actionability | Every step is copy-paste ready: install commands, baseline `lizard src/`, the full threshold invocation `-x"./tests/*" -C 10 -L 100 -a 5` with a flag-meaning table, CSV/XML/HTML output variants, and a CI-fail snippet with explicit `$EXIT` handling — fully executable and covering the common cases per anchor 5. | 5 / 5 |
Workflow Clarity | Steps 1-6 form a coherent install → baseline → thresholds → output → CI-gate → per-language-scope sequence, and Step 5 includes an explicit exit-code checkpoint. However there is no error-recovery loop (e.g., what to do with the report when the gate fails beyond 'see report'), so it matches anchor 4 ('most checkpoints present; minor validation gaps') rather than anchor 5. | 4 / 5 |
Progressive Disclosure | The single file is well-sectioned (When to use, numbered steps, Anti-patterns, Limitations, References) and the only reference — the Lizard README URL — is one level deep and clearly signaled in a References section. No bundle files exist to offload detail, and the ~128-line body is appropriate for the main file; minor gains (e.g., moving the anti-patterns detail out of the main flow) keep it at anchor 4 rather than 5. | 4 / 5 |
Total | 17 / 20 Passed |