Content
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This SKILL.md functions well as a hub/overview document with excellent progressive disclosure through its reference file architecture and clear load-order table. Its main weaknesses are the lack of concrete executable examples within the skill itself (relying entirely on referenced files for implementation detail) and some unnecessary descriptive prose that doesn't add actionable value. The workflow is reasonably clear but lacks validation checkpoints for a system with this many moving parts.
Suggestions
Add a concrete example of the 6-step conversation model in action (even a brief one) so the skill is actionable without loading clinical-spine.md
Add a validation step or checklist for session initialization — e.g., 'Confirm these files loaded: SOUL.md ✓, clinical-spine.md ✓, traditions-quickref.md ✓' to improve workflow clarity
Trim the North Star quote and 'What Satori Does Not Do' section — these are identity/soul content that belongs in SOUL.md, not the operational skill file
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is reasonably well-organized but includes some unnecessary elaboration. The 'What Satori Does Not Do' section and 'North Star' quote add flavor but could be tighter. The opening conventions section is useful but slightly verbose. Some lines like 'A warm, engaged companion who thinks with clinical discipline and speaks with humane, conversational ease' are descriptive padding. | 2 / 3 |
Actionability | The skill provides concrete guidance through the load order table, slash commands, conversation model steps, and opening conventions — these are actionable. However, the actual execution details are entirely delegated to reference files, so this SKILL.md alone doesn't contain executable instructions. The core conversation model is described at a summary level without concrete examples of what good responses look like. | 2 / 3 |
Workflow Clarity | The load order table provides a clear conditional workflow for which files to load when, and the core conversation model outlines a 6-step sequence. However, there are no validation checkpoints — no way to verify correct file loading, no feedback loops for when the conversation model isn't working, and no explicit error recovery steps. The opening conventions provide good branching logic but lack verification. | 2 / 3 |
Progressive Disclosure | This is the skill's strongest dimension. The content is structured as a clear overview/hub that points to well-organized, one-level-deep reference files with explicit load conditions. The table format for file references with 'When to Load' conditions is excellent progressive disclosure. Content is appropriately split between the overview and detailed reference files. | 3 / 3 |
Total | 9 / 12 Passed |