Mostly efficient with executable code throughout, but the Overview's promotional doc-quote bullets ('strong focus on ease of use', 'Interactive terminal based UI', 'Parallel and fast execution') and the near-duplication between the 'How to use' summary and Steps 1-7 add tokens Claude doesn't need.

Provides fully executable commands and copy-paste-ready blocks - install/run, ini and toml config, browse/results/show, pragma examples, mutmut apply + git checkout revert, and a complete CI YAML with an artifact upload step.

A clearly sequenced multi-step process (How to use + Steps 1-7 + worked example) with explicit validation checkpoints - suite green before mutating, new test should fail on the applied mutant - and a feedback loop (apply -> run test -> revert) for the disk-writing operation.

SKILL.md holds the core workflow with a single clearly signaled, one-level-deep reference to references/mutmut-pitfalls.md (a real 1313-byte file) for anti-patterns and limitations, matching the well-signaled one-level-deep anchor.

Lists multiple concrete actions with named commands - 'pip install mutmut', 'runs via mutmut run', 'browses results via mutmut browse or mutmut results', 'applies surviving mutants to disk via mutmut apply {id}', 'suppresses with # pragma: no mutate' - in third-person voice.

Explicitly answers both what (configure/run/browse/apply/suppress mutmut) and when via the explicit trigger clause 'Use for Python codebases needing mutation-quality verification of pytest / unittest suites.'

Uses natural terms a user needing this skill would say - 'mutation testing', 'Python', 'pytest / unittest', 'mutation-quality verification' - matching the rubric's good-coverage anchor rather than the partial-keyword level below.

Names a specific tool (mutmut) and a narrow niche (Python mutation testing) unlikely to conflict with per-language sibling tools like stryker or pitest.