Content
42%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides a comprehensive overview of AI SEO concepts with useful frameworks (three pillars, audit checklist, content type priorities) and specific research-backed metrics. However, it is significantly over-long for a SKILL.md, inlining reference material that should be in separate files, and lacks concrete executable guidance — it tells Claude what to optimize but not how to execute the optimizations step by step. The content reads more like a knowledge base article than an actionable skill.
Suggestions
Move the detailed tables (platform landscape, monitoring tools, Princeton GEO research, content citation shares) into the referenced files like references/platform-ranking-factors.md and references/content-patterns.md, keeping only the top 3-5 most important points inline.
Add concrete, executable examples: instead of 'Add statistics with sources,' show a before/after of a paragraph being optimized, or provide a specific prompt template for rewriting content for AI extractability.
Add a validation/feedback loop after the optimization steps: e.g., 'After optimizing, re-run queries through ChatGPT/Perplexity to verify citation improvement. If not cited after 2-4 weeks, check [specific diagnostic steps].'
Remove sections that explain concepts Claude already knows (what E-E-A-T is, what schema markup does, how different AI platforms work at a basic level) and replace with only the non-obvious, skill-specific insights.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is significantly verbose at ~400+ lines. It explains concepts Claude already knows (how AI search works, what schema markup is, what E-E-A-T means), includes extensive tables that could be in reference files, and repeats information across sections. The 'How AI Search Works' section and 'Common Mistakes' list contain substantial padding. Much of this content (platform landscape tables, monitoring tool comparisons, citation share percentages) should be in referenced files rather than inline. | 2 / 5 |
Actionability | The skill provides structured audit checklists and optimization frameworks with specific metrics (e.g., '40-60 words for snippet extraction'), but lacks executable code, specific commands, or copy-paste ready implementations. The guidance is mostly strategic direction ('Add statistics with sources', 'Ensure your Wikipedia page is accurate') rather than concrete executable steps. The audit tables are useful templates but the actual optimization advice remains at the 'what to do' level rather than 'how to do it' level. | 3 / 5 |
Workflow Clarity | The audit section (Steps 1-4) provides a reasonable sequence, but lacks validation checkpoints or feedback loops. There's no 'verify your changes improved AI visibility' step after optimization, no iterative process for testing and refining, and no clear decision tree for prioritizing which optimizations to tackle first. The workflow is more of a checklist than a guided process with error recovery. | 3 / 5 |
Progressive Disclosure | The skill references two external files (references/platform-ranking-factors.md and references/content-patterns.md) and related skills, which is good structure. However, no bundle files were provided, so these references may be broken. More critically, the skill inlines enormous amounts of content (monitoring tools tables, content type optimization guides, the full Princeton GEO research table) that should be in reference files. The main file tries to be both overview and comprehensive reference, undermining the progressive disclosure pattern. | 3 / 5 |
Total | 11 / 20 Passed |