Content
61%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 skill body is a thorough, largely actionable E-E-A-T assessment framework with concrete scoring rubrics and a ready-to-fill output template. Its main weaknesses are verbosity in places that restate general knowledge, an implicit workflow without explicit validation checkpoints, and a monolithic structure that bundles reference-grade material inline.
Suggestions
Trim sections that restate general knowledge (the AI-citability preamble, basic readability guidance, and generic 'low-quality AI content' phrasing lists) to improve token efficiency.
Add explicit verification checkpoints to the workflow, e.g. 'Confirm each fetched page matches the target URL' and 'Reconcile per-page scores against the E-E-A-T subtotal before writing the report.'
Move the large output template and the detailed point-value scoring tables into reference files (e.g. OUTPUT_TEMPLATE.md, EEAT_RUBRIC.md) and link to them, keeping SKILL.md as a lean overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body runs ~345 lines and, while much is genuine domain-specific scoring rubric, several sections restate widely known ideas (e.g. explaining that AI platforms 'do not just find content — they evaluate whether content deserves to be cited', readability basics, and generic AI-content phrasing lists). It is mostly efficient but includes padding that could be trimmed, so 3 rather than 4. | 3 / 5 |
Actionability | Provides concrete, point-valued scoring tables per E-E-A-T dimension, explicit 'How to Score' instructions, word-count benchmarks, and a full output template the model can fill in. Guidance is mostly executable with minor gaps (e.g. readability is 'estimated without a tool' rather than computed), so 4 rather than 5. | 4 / 5 |
Workflow Clarity | The 'How to Use This Skill' section lists a numbered 1–6 sequence, but checkpoints are implicit and there is no validate/retry feedback loop. Although this is a non-destructive assessment skill, the sequence still lacks explicit verification steps (e.g. confirming fetched pages, cross-checking scores), placing it at 3. | 3 / 5 |
Progressive Disclosure | The body is a single file with no bundle files, but it is well organized into clearly headed sections (E-E-A-T, Content Quality, AI Content, Freshness, Topical Authority, Scoring, Output). Structure is good and easy to navigate; it is not a 5 because the long scoring tables and output template could arguably live in separate reference files, and there are no signaled external references. | 4 / 5 |
Total | 14 / 20 Passed |