Content
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured strategic skill that clearly articulates a 5-layer framework for AI search optimization. Its main strengths are clear organization, a useful failure-patterns section, and appropriate scoping (with explicit 'when NOT to use' boundaries). Its main weaknesses are the lack of concrete, executable examples (no schema markup snippets, no llms.txt templates, no audit commands) and missing bundle files that the skill references.
Suggestions
Add concrete code examples: include a sample llms.txt file, example schema.org JSON-LD snippets for Article/FAQPage/HowTo, and a before/after HTML example showing extraction-friendly content patterns.
Add validation checkpoints to the workflow: specify what tools to use for auditing (e.g., specific schema validators, specific AI products to query), what a passing score looks like for each layer, and what to do when re-testing reveals regressions.
Provide the referenced bundle files (references/llms-txt-guide.md and references/extraction-friendly-patterns.md) or inline the most critical content from them to ensure the skill is self-contained enough to be actionable.
Trim explanatory asides that Claude already understands (e.g., 'AI struggles when claims are spread across paragraphs', 'Mirrors how people prompt AI') to improve token efficiency.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is mostly efficient and well-organized, but includes some sections that could be tightened. For example, the 'When to use' and 'When NOT to use' sections repeat information already conveyed in the description, and some bullet points contain explanatory context Claude doesn't need (e.g., 'AI struggles when claims are spread across paragraphs'). However, it avoids egregious over-explanation and most content earns its place. | 2 / 3 |
Actionability | The skill provides a solid conceptual framework and clear principles, but lacks concrete executable examples. There are no code snippets for schema markup, no example llms.txt content, no sample HTML patterns, and no specific commands or tools for auditing. The guidance is specific enough to act on but falls short of copy-paste ready implementation. | 2 / 3 |
Workflow Clarity | The workflow section provides a clear 6-step sequence, but lacks explicit validation checkpoints and feedback loops. Steps like 'Audit current state' and 'Test current AI visibility' are mentioned but without concrete validation criteria or tools. The 're-test quarterly' note is good but the workflow doesn't specify what constitutes success or failure at each step, nor does it include error recovery guidance. | 2 / 3 |
Progressive Disclosure | The skill references two supporting files (llms-txt-guide.md and extraction-friendly-patterns.md) which is good progressive disclosure design, but no bundle files were provided, meaning these references are broken. The main content is well-structured with clear sections, but the 5-layer framework contains substantial inline detail that could potentially be split into reference files for each layer. The organization is decent but the missing bundle files undermine the disclosure strategy. | 2 / 3 |
Total | 8 / 12 Passed |