Conciseness

The body is lean and assumes competence: "Repowise scores **every file 1–10** from deterministic markers — McCabe complexity, deep nesting, brain methods, class cohesion (LCOM4), god classes, clone detection..." names markers without explaining them, and even the interpretive line about weights being "calibrated against a real defect corpus" is decision-relevant rather than padded. No section explains concepts Claude already knows, so it sits at anchor 5 rather than anchor 4 ('minor instances of over-explanation').

Actionability

Guidance is fully executable with exact call signatures and commands: `get_health(targets=["src/x.py", "src/y.py"])`, literal include-flag values ("biomarkers", "refactoring", "coverage", "trend"), `repowise coverage add cov.lcov` with accepted formats, and `repowise health --recompute`. This matches anchor 5 ('copy-paste ready; specific examples cover the common cases'); anchor 4 would require minor gaps in executable detail, and none are present for the documented surface.

Workflow Clarity

Mode selection is driven by input (dashboard vs. targeted), and the results workflow includes an explicit validation checkpoint — "Check `unresolved` before calling a file clean" — plus feedback loops: "`not_indexed` means run `repowise update`" and no-repository → suggest `/prompts:repowise-init`. The task is read-only analysis, so the destructive/batch cap does not apply; explicit validation and error recovery place it at anchor 5 rather than anchor 4.

Progressive Disclosure

This is a single-file skill (~46 content lines, no references/, scripts/, or assets/ directories exist) whose small API surface needs no external files; the section structure — "Pick the mode by what you pass", "Useful `include` flags", "How to use the results", "CLI equivalents", "Error handling" — is well-organized and easy to navigate. Per the rubric's simple-skill note, this earns anchor 5 with well-organized sections and no buried or nested references.

Specificity

The description richly names the domain — "code health, code quality, complexity, technical debt, risky or hard-to-maintain files, what to refactor next, untested hotspots, or coverage gaps" — with "before/after health reads" as the only concrete action-like phrase, but it never states what the skill actually does (e.g., scoring files, ranking fixes). This exceeds anchor 2 ('names the domain but actions are minimal or generic') because the domain enumeration is far more concrete than 'Processes PDF files', yet it falls short of anchor 4, which requires several explicitly listed actions.

Completeness

The 'when' is explicit and thorough ("Use when the user asks about...", "Also use for before/after health reads..."), but the 'what' is never explicitly stated — capabilities like per-file 1–10 scoring or fix-first ranking appear only in the body, not the description. This mirrors anchor 3's gap level (one half clear, the other missing/implied): it is above anchor 2 ('only when present without what') because the detailed trigger enumeration conveys the domain strongly, but below anchor 4, which requires both 'what' and 'when' to be present.

Trigger Term Quality

Comprehensive natural trigger phrases users would actually say: "code health", "code quality", "complexity", "technical debt", "what to refactor next", "untested hotspots", "coverage gaps", and "planning or finishing a refactor", plus the environment signal "(.repowise/ directory exists)". Multiple synonyms are covered (health/quality/complexity/maintenance risk), matching anchor 5; anchor 4 would require clearly missing natural terms, and none are absent (file extensions don't apply to this non-file-format domain).

Distinctiveness Conflict Risk

The scope gate "in a Repowise-indexed codebase (.repowise/ directory exists)" establishes a clear niche, but generic trigger terms like "code quality" and "complexity" create minor overlap risk with general code-review or refactoring skills when a repo happens to be indexed. This matches anchor 4 ('mostly distinct; minor overlap risk with closely related skills') rather than anchor 5, whose example carries no such broadly shared terms.