Content
68%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 body is highly actionable with concrete checklists and output templates and is well-structured, but it repeats editorial caveats and, as a batch/destructive-oriented skill, lacks explicit validation checkpoints in its workflow. Progressive disclosure is strong structurally though referenced bundle files are absent.
Suggestions
Consolidate the repeated 'not a ranking factor / editorial use only / no authorship inference' caveats into one clearly labeled note instead of restating them in the intro, Step 3, and JSON schema.
Add explicit validation checkpoints to the workflow — e.g., verify extracted statistics trace to sources before scoring E-E-A-T source fidelity, and require a confirmation/preview step before the auto-applying rewrite command in batch mode.
Either include the referenced references/*.md and scripts/cognitive_load.py bundle files so navigation targets resolve, or mark them as optional/external so the dangling paths don't read as broken references.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Content is mostly efficient with table-driven scoring rules, but repeats the same 'not a Google ranking factor / editorial use only / does not infer authorship' caveats across the intro, Step 3, and the JSON schema, and includes some E-E-A-T prose that could be tightened. | 3 / 5 |
Actionability | Provides concrete per-category checklists with points and pass criteria, exact flags, an explicit cognitive_load command, and copy-paste-ready markdown/JSON/table output templates covering the common cases. | 5 / 5 |
Workflow Clarity | The 5-step sequence is clearly numbered, but this is a batch/destructive skill (--batch audits all files; the rewrite command auto-applies changes) with no explicit validation or verification checkpoints, which caps workflow clarity at 3. | 3 / 5 |
Progressive Disclosure | Good section structure with five one-level-deep reference paths listed up front and scoring detail appropriately referenced out, but the referenced references/ and scripts/ paths do not resolve against the actual (empty) bundle, leaving minor navigation gaps. | 4 / 5 |
Total | 15 / 20 Passed |