Lists multiple specific concrete actions: 'unreachable branches, uncalled functions, unused classes, dead feature flags' - these are precise, actionable capabilities.

Clearly answers both what ('Finds and safely removes code that is never executed') and when ('Use when cleaning up after a feature removal, when the user suspects the codebase has accumulated cruft, or when reducing build/bundle size').

Includes natural terms users would say: 'feature removal', 'cruft', 'build/bundle size', 'dead code', 'unused' - these match how developers naturally describe cleanup tasks.

Clear niche focused specifically on dead code removal with distinct triggers like 'unreachable branches', 'dead feature flags', and 'cruft' - unlikely to conflict with general refactoring or code review skills.

Every section earns its place—the deadness classification table, uncertainty checklist, and worked example all provide information Claude wouldn't inherently know about codebase-specific patterns. No padding or explanation of obvious concepts.

Provides concrete verification steps, specific grep commands (`rg 'legacy_format' --type py`), exact file locations to check (`setup.py`, `pyproject.toml`), and a complete worked example with real decision-making. The output format is copy-paste ready.

Clear 5-step workflow (Identify → Verify → Deprecate → Delete → Test) with explicit validation checkpoints. The 'Deprecate first' step for uncertain cases provides a feedback loop, and the worked example demonstrates the verification sequence in practice.

Content is well-organized with clear sections and tables, but it's a moderately long single file. The reference to `behavior-preservation-checker` suggests related skills exist but isn't a clear navigation link. Could benefit from splitting the worked example or detailed tables into separate reference files.