Content
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill delivers highly actionable, well-sequenced coaching guidance with clear branching and concrete scripts. Its main weaknesses are overall length with some cross-branch repetition and a dangling reference to a non-existent examples file.
Suggestions
Move the four full example recommendation outputs into the referenced "examples/" bundle (and create that file) so SKILL.md keeps concise per-branch move lists and points to the long-form examples.
Tighten the recurring Hero Syndrome and 1-on-1 reframe language across branches by defining each pattern once in Key Concepts and referencing it, reducing redundancy.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body (~340 lines) is mostly valuable specialized coaching content Claude would not already know, but the four full prose example-output blocks and repeated framings (Hero Syndrome and the 1-on-1 reframe recur across branches) add length that could be tightened, matching the "mostly efficient but could be tightened" anchor. | 2 / 3 |
Actionability | Recommendations are concrete and ready to deliver — e.g., "Run a Cascading Context Map this week," "Reframe your 1-on-1s immediately" with copy-paste scripts like "I want to understand how each of your products connects to the business goals..." — meeting the actionable standard for an instruction-only skill. | 3 / 3 |
Workflow Clarity | The flow is explicitly sequenced (Step 0 entry mode, Q1/3 routing, then per-branch Q2/3 and Q3/3 with numbered options and "Other (specify)"), with progress labels and a "choose a number to go deeper" follow-up loop; no destructive/batch validation cap applies to a coaching skill. | 3 / 3 |
Progressive Disclosure | Cross-skill references to the mental model and facilitation protocol are clearly signaled and one level deep (a strength), but the body is a long monolithic block with four full inline example outputs that could be externalized, and the referenced "examples/conversation-flow.md" has no corresponding bundle file, fitting the score-2 anchor. | 2 / 3 |
Total | 10 / 12 Passed |