Content
82%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.
The content is a strong, highly actionable workflow with executable commands, a scoring rubric, and concrete output templates, well-structured across four clear phases. Its main weaknesses are minor conciseness redundancy and the absence of explicit validation checkpoints in the analysis flow.
Suggestions
Fix the no-op duplicate on line 28 (`cat AGENTS.md 2>/dev/null || cat AGENTS.md 2>/dev/null`) — both sides are identical so the fallback does nothing; replace the second with a real fallback or remove the `||`.
Add one explicit validation checkpoint in Phase 1 (e.g., 'If fewer than 3 signal categories return results, note the gaps before scoring') to give the analysis flow a feedback loop.
Consider moving the 'Stack reference (internal)' block (lines 175-193) into a `references/stacks.md` file referenced one level deep, to further trim the main body and improve progressive disclosure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly lean — concrete bash commands, a scoring table, and templates with minimal concept explanation — but contains minor trimmable redundancy such as the duplicate `cat AGENTS.md 2>/dev/null || cat AGENTS.md 2>/dev/null` and a long question bank, keeping it just below fully lean. | 4 / 5 |
Actionability | Phase 1 provides copy-paste-ready executable bash commands across six analysis categories, Phase 2 gives a concrete scoring table, and Phases 3-4 supply exact question wording and a fill-in output template, fully covering the common cases. | 5 / 5 |
Workflow Clarity | The four-phase sequence (silent analysis → score stacks → ask 3 questions → recommend) is clearly ordered with explicit constraints ('No more' than 3 questions), but there are no explicit validation checkpoints or feedback loops, which is a minor gap rather than a destructive-operation cap. | 4 / 5 |
Progressive Disclosure | No bundle files exist, so the skill is self-contained; the body is well-organized with clear section headers (Phases 1-4, Stack reference) and the inline stack-reference table is short enough to belong in SKILL.md, with only minor organization gaps. | 4 / 5 |
Total | 17 / 20 Passed |