Content
78%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 lean, well-organized instruction-only skill with a clear workflow and output structure that wastes no tokens. Its main weakness is actionability: it specifies what to audit in concrete terms but provides no commands, tools, or procedures for how to inspect rendered HTML, verify structured data, or check robots directives.
Suggestions
Add concrete inspection methods to the workflow steps — e.g., a curl or tool command for fetching rendered HTML, a specific structured-data validator, and a way to check robots directives and redirects — so each step is executable rather than descriptive.
Define a severity or impact rubric for the 'prioritized fix list' (e.g., how to classify critical vs. minor findings) so the deliverable step produces consistent output.
Include one worked mini-example of an output finding (e.g., a weak title tag with a concrete suggested rewrite) to anchor the 'concrete edits' promised in step 7.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~37-line body assumes Claude's competence: no explanation of what SEO is, why metadata matters, or how search engines work. Every section (Overview, Workflow, Output Pattern, Example Prompts) directs behavior rather than teaching background, matching the 'lean and efficient; every token earns its place' anchor. There is no padding to trim, so 4 or below is ruled out. | 5 / 5 |
Actionability | The workflow names concrete audit items ("Title, meta description, H1/H2 structure, URL, canonical, alt text, schema, and internal links") and a defined output structure, which is genuine concrete guidance — but it gives no how: no commands, tools, or methods for fetching rendered HTML, checking robots directives, or validating structured data. This matches 'some concrete guidance but incomplete; missing key details'. It is above 2 because the checklists and output pattern are specific, not high-level hints, but below 4 because none of the steps are executable as written. | 3 / 5 |
Workflow Clarity | Seven clearly sequenced steps run from target identification through inspection, on-page and technical audits, AI-search and competitor analysis, to a defined deliverable, plus an explicit output pattern — a clear sequence with only a minor gap (no checkpoint such as confirming rendered HTML was actually retrieved before auditing). It is below 5 because there are no explicit validation steps, and above 3 because the sequence is complete, ordered, and specific rather than merely listed with implicit checkpoints. The destructive/batch cap does not apply since this is a read-only audit. | 4 / 5 |
Progressive Disclosure | The skill is under 50 lines with no external references needed, and content is organized into four well-signaled sections — per the rubric guideline, this earns a 5 on well-organized sections alone. There are no buried or nested references, and no inlined content that belongs in a separate file, ruling out lower anchors. | 5 / 5 |
Total | 17 / 20 Passed |