Content
53%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 provides a well-structured eligibility/scoring framework and clear principles, but its JSON-LD example is a stub, validation steps lack runnable commands and an explicit error-recovery loop, and some sections contain boilerplate and over-explanation. Tightening the body and making the examples executable would raise the score.
Suggestions
Replace the stub JSON-LD block with a complete, copy-paste-ready example (e.g. a real Article or Product block) covering the common cases.
Add an explicit validate→fix→retry feedback loop with runnable tool references (e.g. Google Rich Results Test URL / CLI) so deployment failures have a recovery path.
Trim the boilerplate "When to Use" line and the empty '---' section, and consider moving the full schema-type catalog and scoring rubric into a reference file to improve conciseness and progressive disclosure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient structured lists, but the supported-types catalog, generic "Use for:" prose, a duplicate empty '---' section (line 9), and a boilerplate "When to Use" line (line 360) add explanation and noise that could be trimmed. | 3 / 5 |
Actionability | Concrete guidance exists (scoring rubric with weights and bands, supported types, validation checklist, output format), but the JSON-LD example is a non-executable stub ({"@type": "..."}) and validation tools are named without commands, leaving key details incomplete. | 3 / 5 |
Workflow Clarity | Phases 0→1 are sequenced with a score threshold (≥70) and a stop rule for "Do Not Implement", but there is no explicit validate→fix→retry feedback loop with runnable commands, which the rubric treats as a checkpoint gap. | 3 / 5 |
Progressive Disclosure | No bundle files exist, so all content lives in a single SKILL.md; it is well-organized with clear section headers and no nested references, though the ~360-line body (including the inline scoring rubric and type catalog) could be partially split into reference files. | 4 / 5 |
Total | 13 / 20 Passed |