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: lean, concrete, and immediately executable, with a clear workflow, checklist, error-recovery path, and a worked example. The only structural note is that detailed tool output examples are all inline, which is acceptable at this size but worth splitting if the skill expands.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~95-line body is lean with no explanations of concepts Claude already knows: terse sections ("## Workflow", "## Checklist"), compact risk tables, and a worked example that adds concrete grounding the abstract workflow lacks. Every token earns its place. | 5 / 5 |
Actionability | Tool calls are fully concrete with parameters (impact({target, direction: "upstream", minConfidence: 0.8, maxDepth: 3})), sample outputs with real symbol names and confidence values, and an exact terminal recovery command (node .gitnexus/run.cjs analyze). | 5 / 5 |
Workflow Clarity | The 4-step numbered workflow, checklist, and explicit error-recovery feedback loop ("If 'Index is stale' → run node .gitnexus/run.cjs analyze") sequence the process clearly with a reporting checkpoint. This is read-only analysis, so the destructive-operation validation cap does not apply. | 5 / 5 |
Progressive Disclosure | The body is well-sectioned with no nested references and no buried content, appropriate for a standalone skill with no bundle files. Not 5 because the detailed tool-output examples in the Tools section are fully inlined with no external split, a minor organization gap if the skill grows. | 4 / 5 |
Total | 19 / 20 Passed |