Content
43%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-sectioned catalogue stub, but it provides almost no actionable guidance: the real workflow lives entirely in an upstream GitHub repo and the single instruction reduces to "go inspect the upstream README". As a SKILL.md it reads as an advertisement rather than an executable skill, leaving both actionability and workflow clarity well below par.
Suggestions
Replace the `open https://github.com/figma/skills` placeholder with a concrete, executable install step (e.g., a git clone or copy command with the exact target path in the agent's skills directory).
Inline or bundle the core workflow (how to generate the design system rules: what inputs to read from Figma, what rule format to emit, where to write it) in SKILL.md or a clearly referenced one-level-deep file, instead of deferring all content to an external repo.
Add a short verification step so the agent can confirm the skill is installed and invocable (e.g., check the trigger phrases resolve to the installed bundle) before telling the user it is ready.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is short and lean with no explanation of concepts Claude already knows, matching anchor 4 (efficient with minor trimming possible). It is not a 5 because the frontmatter description is repeated verbatim in "What it does" and source/category data from the frontmatter is restated in the body. | 4 / 5 |
Actionability | The only "command" is `open https://github.com/figma/skills` alongside "Inspect the upstream README for exact paths" — a high-level hint with no concrete install command or executable steps, matching anchor 2. It is not 3 because there is no partially-executable guidance, just a pointer to go read an external README. | 2 / 5 |
Workflow Clarity | A rough sequence exists (inspect upstream, install the bundle, invoke by name or trigger), but the critical install step is entirely undefined and there are no checkpoints, matching anchor 2. It does not reach 3 because the listed steps have a major gap rather than merely missing validation, and the simple-skill exception does not apply since the single action (installing and running the skill) is ambiguous. | 2 / 5 |
Progressive Disclosure | No bundle files (references/, scripts/, assets/) exist, and the sections are clearly labeled, but the actual skill content lives in an out-of-band GitHub URL rather than clearly signaled, one-level-deep reference files — matching anchor 3 (some structure but could be better organized). It is not 4/5 because the substantive content the skill advertises is not present or navigable locally. | 3 / 5 |
Total | 11 / 20 Passed |