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.
A tight, highly actionable seven-phase workflow with concrete commands, output schemas, and explicit filtering/ranking rules. Its only notable gaps are a missing error-recovery feedback loop and a cross-skill asset dependency, which keep workflow clarity and progressive disclosure at 4 rather than 5.
Suggestions
Add an explicit validate→fix→retry loop (e.g., re-run profiling or re-scan a module when a hit matrix cell is ambiguous) to push workflow clarity toward 5.
Trim the motivational sentences in Phase 3 ('This history teaches you what to look for. Successful past techniques are strong signals for where to look next.') to lift conciseness.
Note the cross-skill dependency on lading-optimize-hunt/assets/db.yaml explicitly (or vendor a local copy) so the skill is self-contained and progressive disclosure is unambiguous.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and mostly assumes Claude's competence — a 14-row pattern table, terse phase descriptions, and concrete output formats with little padding; mild over-explanation ('This history teaches you what to look for. Successful past techniques are strong signals...') keeps it just short of the lean 5 anchor. | 4 / 5 |
Actionability | Provides copy-paste-ready guidance throughout — explicit Glob patterns, a runnable script path (scripts/profile-modules, verified present), the TSV output schema, a required hit-matrix format, opportunity tuples, ranking rules, and a final YAML template — matching 'fully executable; specific examples cover the common cases'. | 5 / 5 |
Workflow Clarity | Seven clearly sequenced phases with a 'must follow exactly' procedure, an explicit STOP gate in Phase 5, and a deterministic tiebreaker; it avoids the batch-operation cap because Phase 5 acts as a validation gate, but lacks an explicit validate→fix→retry feedback loop that would reach the 5 anchor. | 4 / 5 |
Progressive Disclosure | Well-organized into seven headed phases with one-level-deep, clearly signaled references (the bundled scripts/profile-modules and the cross-skill db.yaml path); it is not the simple-skill 5 case (~120 lines, external references) and the cross-skill dependency is a minor organization note rather than a monolith. | 4 / 5 |
Total | 17 / 20 Passed |