Content
50%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 is actionable with a ready-to-use CLI example, but it is noticeably verbose due to duplicated input descriptions and fully inlined JSON schemas, and lacks validation checkpoints in its workflow.
Suggestions
Remove the duplicated sceneRef description (it appears in both the 'Inputs' section and the JSON schema) to cut redundancy.
Add a validation/verification step in the Behavior or a workflow section, e.g., confirm the returned snapshot shows the target scene as active before reporting success.
Consider moving the full input/output JSON schemas into a separate reference file and keeping only the essential shape inline, to improve token efficiency and progressive disclosure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body duplicates the description and repeats the sceneRef input description verbatim in both the Inputs section and the schema, and inlines large JSON schemas that a competent Claude does not need fully spelled out, making it noticeably verbose. | 2 / 5 |
Actionability | Provides a concrete, copy-paste-ready CLI invocation plus file/stdin alternatives and troubleshooting, with only minor gaps (the example uses a placeholder 'string_value' rather than a realistic sceneRef object). | 4 / 5 |
Workflow Clarity | Behavior describes the resolution sequence, but there is no validation checkpoint confirming the scene was actually set active or handling the case where the named scene is not among opened scenes, leaving the workflow at the 3-anchor level. | 3 / 5 |
Progressive Disclosure | There are no bundle files; content is organized into sections but the large input/output JSON schemas are inlined entirely in SKILL.md rather than split out, which is structure that could be better organized. | 3 / 5 |
Total | 12 / 20 Passed |