finding-dead-code

Dead code detection requires verifying exact callers and demanding concrete usage evidence before marking items as alive. This approach scans the entire codebase, including tests, to distinguish truly unused code from active items and reduce false positives.

Transitive dead code consists of functions or modules only called by other already-dead code. Identification groups these orphaned items together by tracing the call graph, ensuring that cascading unused code is safely detected and planned for removal.

Static analysis can detect write-only dead code by identifying variables, setters, or storages that are updated but never read. This requires thorough provenance tracking across the repository to prove no read paths exist.

Safe removal involves generating an evidence-rich report with a deletion plan and post-removal verification commands. Enforcing Git safety and performing iterative re-scans ensures dependencies remain intact after the dead code is deleted.

Dead code analysis applies across the entire repository, including test and auxiliary code. Scanning these directories is essential because items appearing unused in production logic may actually be invoked by test suites.