Content
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is concise and well-structured as a catalogue pointer, but it is not actionable: it defers the real workflow to an external upstream README that is not bundled, offering only a URL-open command.
Suggestions
Provide concrete install/invocation steps (e.g., exact clone path and target skills directory) instead of deferring entirely to the upstream README.
Include a minimal inline example of the code-to-Figma workflow or a direct link to the specific upstream reference file so the skill is usable without external lookup.
Add a short verification step confirming the skill is installed and invokable before directing the agent to trigger it.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and sectioned ('What it does', 'Source', 'How to use') with no padding of concepts Claude already knows, so every token earns its place rather than the score-2 mildly-padded anchor. | 3 / 3 |
Actionability | The only concrete command is `open https://github.com/figma/skills`, and the actual task is deferred to an absent upstream README ('Inspect the upstream README for exact paths'), leaving no executable guidance for the skill's purpose; it does not reach 2 because no concrete task steps are provided. | 1 / 3 |
Workflow Clarity | A loose sequence exists (inspect README, install bundle, invoke by name/triggers) but the critical 'install the upstream bundle' step is unspecified, so it sits above the no-sequence anchor yet below the explicit-checkpoint anchor. | 2 / 3 |
Progressive Disclosure | With no bundle files present and the body under 50 lines, the well-organized sections plus a single clearly-signaled upstream reference satisfy the simple-skill anchor; it is not a 2 because content is appropriately split and navigable. | 3 / 3 |
Total | 9 / 12 Passed |