Content
65%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is highly actionable with executable commands and concrete schemas, but it is held back by triplicate input descriptions, an inlined output schema, and an implicit rather than explicit validation/recovery workflow.
Suggestions
Describe the inputs once (keep the table or the schema, not both the "Inputs" list and the "Input" table and the schema) to remove redundancy.
Move the full EditorStatsData output schema into a references/ file and link to it, keeping SKILL.md as an overview.
Add a short numbered workflow (get state -> set state -> if it throws on compile errors, fix them and retry) with an explicit validation checkpoint.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly operational and free of concept-explaining fluff, but the two boolean inputs are described three times ("Inputs" list, the "Input" table, and the Input JSON Schema) and the full EditorStatsData output schema is inlined, so it could be tightened. | 2 / 3 |
Actionability | It provides a copy-paste-ready bash command with a complete JSON input plus input-file and stdin variants, and the behavior section states the exact throw condition, so the guidance is fully executable. | 3 / 3 |
Workflow Clarity | The single action is unambiguous and the compile-error throw is mentioned, but there is no explicit get-state -> set-state -> on-throw-fix-and-retry sequence with a validation checkpoint, leaving the workflow implicit. | 2 / 3 |
Progressive Disclosure | Sections are well-organized, but there are no external reference files and the full eight-property output schema is inlined in SKILL.md rather than split out, so content that could be separate is inline. | 2 / 3 |
Total | 9 / 12 Passed |