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 well-built operational skill body: exact runnable commands, a handoff template, explicit thresholds, evidence-labelled flags, and honest limits. The main improvement areas are trimming inline citation detail that duplicates evidence-map.md and tightening reference hygiene (dangling references/papers/ path, unlinked library.md).
Suggestions
Trim the citation parentheticals in the flags table to bare evidence labels (e.g. "research (Frandsen 2025)") since references/evidence-map.md already carries the study details, saving table tokens.
Fix reference hygiene: either add the missing references/papers/ directory (with INDEX.md) the body points to, or link references/library.md from the Papers section so the paper inventory is reachable in one level from SKILL.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes competence — no explanations of concepts Claude already knows — but the flags table carries inline citation parentheticals (e.g. "a 10 % rule did not prevent injuries in an RCT, Buist 2008") whose details already live in references/evidence-map.md and could be trimmed to just the evidence label. | 4 / 5 |
Actionability | Fully executable: copy-paste commands with exact paths and all flags (--json, --today YYYY-MM-DD, --id N, --no-readiness), concrete check-in syntax (athlete.py check-in --area .. --pain N [--gait-change]), a handoff block template, and explicit thresholds with their file location (scripts/analysis.py). | 5 / 5 |
Workflow Clarity | The sequence is present across sections (run scripts → lead with the comparison → flags → one action per flag → ask and record → handoff block) with real checkpoints (gait-change pain means stop and recommend a clinician; verify a paper before quoting effects). It falls short of a 5 because no single explicit end-to-end order ties the sections together. | 4 / 5 |
Progressive Disclosure | Good structure: references/evidence-map.md and scripts/analysis.py are clearly signaled and exist, and content is appropriately split. Minor gaps: the body points to references/papers/ which is absent from the bundle, and references/library.md is never linked from SKILL.md (reachable only as a second level via evidence-map.md). | 4 / 5 |
Total | 17 / 20 Passed |