Content
92%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A well-organized, actionable instruction skill with clear sequencing and validation, scoring highly on conciseness, actionability, and workflow clarity. Its one real defect is a missing referenced file (templates/SOUL.template.md) that the skill depends on but that is absent from the bundle.
Suggestions
Add the missing templates/SOUL.template.md file (or correct the path) — the skill instructs Claude to read it before generating, so its absence breaks the documented workflow.
Keep references to a single navigable level: confirm every path named in the Architecture block (SKILL.md, templates/SOUL.template.md, references/conversation-guide.md) actually exists in the bundle.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean and structured with tables for phases and the extraction tracker; it assumes Claude's competence and never lectures on concepts Claude already knows, so every section earns its place. | 3 / 3 |
Actionability | Concrete, executable guidance throughout — a precise tool call setup_agent(soul=..., description=...), explicit round budgets (5–8), required-field checklist, and a sharpened example ("argue position, push back, speak truth not comfort" — not "honest and brave"). | 3 / 3 |
Workflow Clarity | The four-phase conversation and the six-step numbered Generation flow are clearly sequenced with validation checkpoints (all required fields before generating, iterate until confirmed) and an explicit error-recovery loop ("If setup_agent returns an error, report it... do not claim success"). | 3 / 3 |
Progressive Disclosure | The overview is well signaled and references/conversation-guide.md is a real one-level-deep file, but the body repeatedly directs Claude to read templates/SOUL.template.md and no templates/ directory exists in the bundle, so navigation breaks for that path. | 2 / 3 |
Total | 11 / 12 Passed |