Content
22%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is an early draft: it states a core idea and one example trigger but provides no concrete questions, workflow, or actionable guidance. It reads as a placeholder rather than a usable skill body.
Suggestions
Replace the draft stub with concrete guidance, e.g. a list of pointed question categories to ask (remaining work, avoided work, importance) and the order to ask them in.
Add a short workflow with an explicit checkpoint, such as summarizing the extracted priorities back to the user for confirmation.
Remove the "DRAFT — being refined" meta-note once content is finalized, as it adds tokens without actionable value.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is short and not padded with concepts Claude already knows, but tokens like "DRAFT — being refined with the user" and the restated core idea do not earn their place as actionable instruction. Not 3 because most tokens convey no usable guidance; not 1 because it is lean rather than verbose. | 2 / 3 |
Actionability | "Core idea: the agent interviews the user to extract priorities..." describes the intent rather than instructing; there is no concrete example question, script, or executable guidance. Not above 1 because it gives Claude nothing concrete to actually do or say. | 1 / 3 |
Workflow Clarity | No steps, sequence, or validation checkpoints are present — the body is an unfinished draft stub. Not above 1 because there is no discernible workflow to follow. | 1 / 3 |
Progressive Disclosure | There are no references or nested files, and for a tiny single-purpose skill that can be acceptable, but the body is not organized into usable sections — it is an unstructured draft. Not 1 because it avoids deep/nested references; not 3 because nothing is organized for navigation or use. | 2 / 3 |
Total | 6 / 12 Passed |