Content
96%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.
An exemplary skill body: concise spec tables, executable JSON examples for every node type, workflows with explicit validation checkpoints and error-recovery guidance, and a clearly signaled external examples file. The only structural improvement would be splitting the full attribute reference into a separate file to keep SKILL.md as a leaner overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is almost entirely dense attribute tables, JSON snippets, and non-obvious pitfalls ("Use \n for line breaks in JSON strings. Do not use the literal \\n", "Preset color values are intentionally undefined -- applications use their own brand colors") with virtually no padded prose or explanation of concepts Claude already knows. This is niche spec knowledge Claude does not have, so every token earns its place. | 5 / 5 |
Actionability | Every node and edge type has a complete, copy-paste-ready JSON example, and the attribute tables give exact types, defaults, and allowed values (e.g., "fromEnd | No | string | none | none or arrow"). ID generation is specified with concrete samples ("6f0ad84f44ce9c17"), covering the common cases fully. | 5 / 5 |
Workflow Clarity | Four workflows are clearly sequenced, each ending with an explicit bolded **Validate** step, plus an 8-item Validation Checklist and error-recovery guidance ("If validation fails, check for duplicate IDs, dangling edge references, or malformed JSON strings"). This matches the anchor-5 feedback-loop pattern. | 5 / 5 |
Progressive Disclosure | The body is well-sectioned and points to a real, one-level-deep reference ("See [references/EXAMPLES.md](references/EXAMPLES.md) for full canvas examples including mind maps, project boards, research canvases, and flowcharts") that exists and contains no nested references. However, nearly the entire attribute/spec reference is inlined in SKILL.md with only one external file, so bulk reference content could arguably be split out; this sits between anchors 4 and 5. | 4 / 5 |
Total | 19 / 20 Passed |