Conciseness

The body is efficient and operational throughout — no padding, no conceptual lecture, and sections like the missed-mutant example pattern earn their tokens. It falls just short of anchor 5 because of minor over-explanation Claude doesn't need, e.g. the parentheticals defining Caught/Missed ('good; test suite detected the bug', 'a plausible bug went undetected') restate mutation-testing semantics. Not anchor 3: there is no genuinely unnecessary section or flabby explanation.

Actionability

Guidance is fully executable and copy-paste ready: `cargo mutants --version` with an exact install fallback (`cargo install cargo-mutants --locked`), complete commands with flags for both diff and full modes, an explicit empty-diff stop condition, a worked example of how to explain a missed mutant, and a concrete score formula (`<caught> / <total>`). This matches anchor 5 ('fully executable; copy-paste ready commands; specific examples cover the common cases').

Workflow Clarity

The sequence is clear (verify install → choose mode → run → interpret → report) with several real checkpoints: install fallback on failure, empty-diff detection with an explicit stop, waiting for the user's mode choice, and warning about the 15–60 min runtime. It is not anchor 5 because there is no error-recovery loop for the run itself — no guidance for a failed or partially completed cargo-mutants run (e.g. resuming from mutants.out.old, which the command deletes). Anchor 3 would require missing or merely implicit checkpoints, which is not the case.

Progressive Disclosure

No bundle files exist, and the body is well-organized into clearly labeled sections (Choose a mode, Run, Interpret, Output format) that all belong inline at this size. This is good structure with content appropriately placed (anchor 4). It is not anchor 5: at ~68 lines it exceeds the under-50-line simple-skill case where well-organized sections alone earn a 5, and there is no reference signaling to evaluate — the output-format section is the only borderline candidate to split out.

Specificity

The description names the domain and tool concretely ("Run mutation testing with cargo-mutants on the Rust workspace") but lists only one real action — running the tests. The other phrases ("check test quality", "find untested code paths") are framed as trigger conditions rather than distinct capabilities, so coverage is not comprehensive. It fits anchor 3 ('names domain and 1-2 concrete actions, but not comprehensive'): it is well above a generic 'Processes PDF files' (2) but does not enumerate several specific actions like result interpretation, test suggestions, or scoring (4).

Completeness

It explicitly answers both questions: the 'what' is "Run mutation testing with cargo-mutants on the Rust workspace" and the 'when' is the explicit clause "Use when asked to run mutation tests, check test quality, or find untested code paths" with three concrete trigger phrases. This matches the anchor-5 good example pattern (clear what + explicit 'Use when...' triggers), and is clearly better than anchor 4's weaker, less specific 'when'.

Trigger Term Quality

"run mutation tests", "check test quality", and "find untested code paths" are natural phrases a user would say, giving good keyword coverage. It is not a 5 because common synonyms and variations are missing — e.g. "test coverage", "mutation score", or tool-name triggers like "cargo mutants". It is clearly above anchor 3, which only requires 'some relevant keywords' with missing variations.

Distinctiveness Conflict Risk

The niche is distinct (cargo-mutants, Rust workspace, mutation testing) and the triggers ("run mutation tests", "find untested code paths") are specific to this skill, so the risk of firing for the wrong skill is minimal. A generic coverage tool might overlap slightly, but this matches anchor 5 ('clear niche with distinct triggers; minimal conflict risk').