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 body is a lean, well-organized catalogue stub, but as a standalone SKILL.md it provides almost no executable guidance: the actual workflow is delegated to an external GitHub URL with no install steps, and 'What it does' merely duplicates the frontmatter description. It works as an advertisement entry but scores weakly on actionability and workflow clarity.
Suggestions
Replace the 'open https://github.com/...' inspection step with concrete, executable install commands (e.g. git clone of the upstream repo and a copy/symlink into the skills directory) so the agent can act without browsing.
Drop the 'What it does' section that duplicates the frontmatter description verbatim, and instead summarize what the 14 sub-skills cover and when to reach for which.
Add a short validation step after install (e.g. how to confirm the apple-hig skill is discoverable) so the workflow has a checkpoint for error recovery.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is short and lean with no explanation of concepts Claude already knows; the one redundancy is the 'What it does' section repeating the frontmatter description verbatim. Not 5 because that duplicated block does not earn its tokens; not 3 because there is no padding or unnecessary explanation beyond that. | 4 / 5 |
Actionability | The core instruction 'install the upstream bundle into your active agent's skills directory' is a high-level hint with no specific steps or commands; the only concrete command is 'open https://github.com/raintree-technology/apple-hig-skills', which inspects rather than executes anything. Anchor 2 ('minimal concrete guidance; high-level hints but missing the specific steps') fits; not 3 because there is no executable step toward the actual task, and not 1 because a concrete command and a concrete next action (invoke by name) do exist. | 2 / 5 |
Workflow Clarity | A rough sequence is present (inspect upstream README, install bundle, invoke by name or trigger phrase) but the steps are loosely defined and there are no checkpoints or validation. Not 4 because 'install the upstream bundle' has no concrete sub-steps; not 2 because the ordering is coherent and the final invocation step is specific. This is a non-destructive skill, so no hard cap applies. | 3 / 5 |
Progressive Disclosure | No bundle files exist (no references/, scripts/, or assets/), and the body is under 50 lines with well-organized sections ('What it does', 'Source', 'How to use'). Not 5 because all substantive content is deferred to an external URL rather than to clearly signaled one-level-deep bundled references, leaving the skill with no local detail layer; not 3 because what is present is well structured and the deferral is explicitly signaled. | 4 / 5 |
Total | 13 / 20 Passed |